Document of The World Bank Report No: 25713 IMPLEMENTATION COMPLETION REPORT (CPL-36540; SCPD-3654S) ON A LOAN IN THE AMOUNT OF US$22 MILLION TO THE REPUBLIC OF COSTA RICA FOR A HEALTH SECTOR REFORM PROJECT May 14, 2003 Central America Department Country Management Unit Latin America and The Caribbean Region CURRENCY EQUIVALENTS (Exchange Rate Effective May 2003) Currency Unit = Colones 392.91 Colones = US$ 1 US$ .0025 = Colones 1 FISCAL YEAR July 1 June 30 ABBREVIATIONS AND ACRONYMS ABBREVIATIONS AND ACRONYMS CCSS Costa Rica Social Security Agency DRG Diagnostic Related Group EBAIS Basic Team for Comprehensive Health Services GDP Gross Domestic Product ICB International Competitive Bidding INCIENSA Costa Rican Institute of Research and Teaching INEC National Institute of Statistics and Census IPC Index of Consumer Prices IVM Insurance for Disability, Old Age and Death MOF Ministry of Finance MOH Costa Rica Ministry of Health OPC Agency for Complementary Pensions PEA Active Economic Population PHC Primary Health Care SS Social Security UCR University of Costa Rica UPH Unit for Hospital Facilities Production Vice President: David de Ferranti Country Director: Jane Armitage Sector Director: Ana Maria Arriagada Task Team Leader: Maria Luisa Escobar COSTA RICA CR HEALTH SECTOR REFORM CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 3 4. Achievement of Objective and Outputs 6 5. Major Factors Affecting Implementation and Outcome 18 6. Sustainability 19 7. Bank and Borrower Performance 20 8. Lessons Learned 22 9. Partner Comments 27 10. Additional Information 29 Annex 1. Key Performance Indicators/Log Frame Matrix 30 Annex 2. Project Costs and Financing 36 Annex 3. Economic Costs and Benefits 40 Annex 4. Bank Inputs 55 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 57 Annex 6. Ratings of Bank and Borrower Performance 58 Annex 7. List of Supporting Documents 59 Annex 8. Supporting Data and tables 61 Project ID: P006954 Project Name: CR HEALTH SECTOR REFORM Team Leader: Maria-Luisa Escobar TL Unit: LCSHH ICR Type: Core ICR Report Date: May 16, 2003 1. Project Data Name: CR HEALTH SECTOR REFORM L/C/TF Number: CPL-36540; SCPD-3654S Country/Department: COSTA RICA Region: Latin America and Caribbean Region Sector/subsector: Health (90%); Health insurance (5%); Central government administration (5%) Theme: Health system performance (P); Administrative and civil service reform (S); Decentralization (S) KEY DATES Original Revised/Actual PCD: 02/10/1993 Effective: 02/10/1995 04/18/1995 Appraisal: 02/24/1993 MTR: 06/15/1996 07/07/1997 Approval: 10/21/1993 Closing: 01/31/1999 09/30/2002 Borrower/Implementing Agency: GOVERNMENT OF COSTA RICA/CAJA SEGURO SOCIAL Other Partners: STAFF Current At Appraisal Vice President: David de Ferranti Shahid Javed Burki Country Manager: Jane Armitage Donna Dowsett Coirolo Sector Manager: Evangeline Javier Kye Woo Lee Team Leader at ICR: Maria Luisa Escobar Carmen Hamann ICR Primary Author: Maria Luisa Escobar; James Cercone 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome:HS Sustainability:HL Institutional Development Impact:H Bank Performance:HS Borrower Performance:HS QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: No A New Approach to the ICR Health Reform in Costa Rica and Bank Support A new Approach to the ICR In principle, identification of the optimal combination of efforts that generate a desired outcome from the implementation of health reforms should result from lessons learned from Bank support of reform processes in several countries. While this ICR is not an intensive learning ICR, this ICR is designed as a contribution to that exercise. A reform process is a combination of efforts over a period of time including those supported by Bank operations. Project financed activities evolve in an environment determined by factors such as the economic and fiscal situation of the country in question, the political leadership as well as the degree of community participation, the particular characteristics of the health sector, etc. This ICR has been designed with this holistic approach in mind, including stakeholder analysis and follow-up workshops, in order to analyze Bank support to the reform process within the particular characteristics of the sector at the time of project design and during its implementation. This analysis identifies the combination of efforts that produced the observed results in the health reform process in Costa Rica while trying to establish a relationship between the successes and failures of reform processes with the combination of activities and events that interacted with project design and implementation. The ICR preparation process involved the analysis of issues of diverse nature: from the timeliness of political decisions, to economic situations that enable or disable the implementation of strategies designed to be supported by technical assistance, to the overall political and social climate of the country and its institution(s). A detailed analysis of results obtained during the project's life, compared to initial objectives and to project activities has been done, and then those are analyzed within the political and socio-economic context present during the whole project cycle. Structured interviews were done with about 30 people who have been involved in the health reform process in Costa Rica in different capacities. The interview results were compiled and analyzed for the ICR analysis. - 2 - 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The project's stated objective was to support the Government's effort to implement critical policy, institutional and operational reforms which aimed at improving efficiency, effectiveness and quality of the delivery of health services by the Caja Costarricense de Seguro Social (CCSS). The project proposed five main strategies to achieve these objectives: (i) assisting the CCSS in the implementation of institutional reforms at the central level and supporting the decentralization process; (ii) supporting the implementation of a redesigned primary health care model; (iii) strengthening the CCSS financial resource management, resource allocation and pilot testing of alternative models of health care financing; and (iv) supporting the national surveillance and quality control laboratories under the Instituto Nacional de Ciencias en Salud (INCIENSA). The project was expected to contribute to poverty alleviation by improving the access of the poor to good quality health care services. The project's objectives were consistent with the Bank's assistance strategy for Costa Rica, discussed at the Board on March 10, 1993, which was to ensure macroeconomic stability, reform the public sector, integrate the economy with world markets, strengthen infrastructure, enhance competitiveness and efficiency of the financial system, improve coverage and efficiency of social programs to further reduce poverty and improve environmental management. The loan was the first Bank operation to Costa Rica in the health sector and was seen as an important vehicle to strengthen the social sector policy dialogue with the Government. The loan was complementary to a parallel operation by the Interamerican Development Bank (IDB) that focused on strengthening the Ministry of Health, building a regional hospital and providing infrastructure for the integrated primary care model of the CCSS. The project objectives and the proposed investment program were developed as part of a long running dialogue between the Bank and the CCSS. In the early 1990s a number of national committees developed clear objectives to improve primary care and strengthen the MOH and INCIENSA. The project was designed to support the actions that were developed as a part of a national consensus building exercise. Concern about the increasing expenditures in the health sector, approaching 10 percent of GDP at the time of project design, and high levels of evasion of social security contributions, led to the incorporation of specific objectives aimed at improving efficiency and effectiveness of the resource management and allocation systems. Further, the separation of financing and delivery functions was introduced as a means to promote decentralization and generate further efficiency in the health sector. 3.2 Revised Objective: The project supports the Government's effort to implement policy, institutional and operational reforms aiming at improving the efficiency, effectiveness and quality of delivery of CCSS health care services and enhance the quality control and surveillance system in the health sector. This will be accomplished by: a) assisting the CCSS to implement institutional reforms at the central and regional levels; b) implementing a redefined primary health care model; c) strengthening the CCSS financing related to resource management, allocation and mobilization, reduction of premium evasions, and pilot testing of alternative health financing models; and d) reengineering of the - 3 - National Surveillance and Quality Control Laboratory (INCIENSA). 3.3 Original Components: Component I. CCSS Institutional Reform and Development (US$9.1 million or 28% of total project costs). This component was to support technical assistance, specific studies, field surveys, seminars, training and study tours areas through four subcomponents aimed at developing and strengthening: (a) the CCSS organizational structure (US$1.4 million); (b) management information system (US$3.6 million); (c) human resources development and manpower planning (US$1.6 million); and (d) pharmaceutical and medical supplies system and rational use of medicines (US$2.5 million). The main elements that were supported under this component (subcomponents (a) to (c)) were designed to provide the CCSS with a redesigned organizational and functional structure that was oriented to the separation of the pensions, healthcare and financial functions of the CCSS. To this end, the project proposed to include support through management information systems (MIS) and human resource planning and development. Prior to effectiveness, the CCSS provided the Bank with numerous studies and terms of reference that were focused on supporting the component's objectives. A central aspect of this component was the implementation of the Human Resources Development Fund that was to finance the requirements of the CCSS training program for implementing the reforms. The Fund was to have been merit-based to allow decentralized entities to obtain access to training funds. The fourth subcomponent, Pharmaceutical and Medical Supplies, was aimed at improvements in the system of supplies of pharmaceutical and medical materials, and efforts to promote the rational use of medicines by improving prescription and control practices. This component was intended to address the myriad problems with the planning, procurement, distribution and prescription of pharmaceuticals encountered during project design. The proposed measures were to have been implemented during the first two years of the project. Component II. Redefined Primary Health Care Model, Quality Assurance and Hospital Study (US$17.0 million or 53.1% of total costs). This component was to address the organizational and technical reforms needed to improve the coverage, quality and effectiveness of the primary health care services provided by the CCSS. The component was to finance equipment, studies, training, study tours and incremental salaries for staff transferred from the MOH to the CCSS. The component was to be executed through three sub-components: (a) redefined primary health care model (US$16 million); (b) quality assurance (US$0.5 million); and (c) Hospital Study (US$0.5 million). Over 94 percent of the component's costs were allocated to the implementation of the redefined primary care model. This component was to provide critical support for the implementation of primary care in the CCSS. Prior to the project, primary care services were provided mainly through the MOH. As part of the reform process, the MOH transferred all responsibility for direct provision of care to the CCSS, assuming thereafter the functions of regulation and policy making. The component proposed to establish 74 health areas (Areas Medicas) and 700 basic health teams (Equipo Basico de Atencion Integral de la Salud, EBAIS) as the central strategy to making health services more responsive to community needs and to improving access to a redesigned model of primary care services. The remaining six percent of the component's costs were assigned to support technical assistance for implementation of a quality assurance system and to the analysis and design of reforms for the hospital sector. The quality assurance aspect was focused on the development of indicators and quality assurance mechanisms in support of the PHC model. The hospital study proposed to establish a blueprint, similar to that which was prepared for the PHC model, as the basis for the implementation of broader hospital reform - 4 - measures. Component III. Resource Management and Pilot Testing of Alternative Models of Health Care Financing (US$2.3 million or 7.2 percent of total project costs). This component was to finance studies, information systems and pilot testing of alternative financing models to allocate and transfer resources to all levels of care. The component consisted of four sub-components: (a) resource management, allocation, and evaluation (US$0.9 million); (b) resource mobilization and improvement of the financial sustainability of the CCSS (US$0.3 million); (c) review of existing alternative resource allocation models and alternative delivery and financing models (US$0.6 million); and (d) pilot testing of alternative financing and administration mechanisms (US$0.5 million). The component focused on providing support for the two main areas related to financial management: revenue generation and management and resource allocation. With regard to resource allocation, the component was to support the introduction of mechanisms oriented towards replacing the historical budgeting system with a performance-based resource allocation system, which would match expenditures against program outcomes and institutional objectives. Three of the four subcomponents were focused on carrying out studies, designing and testing resource allocation mechanisms. Areas that were targeted include the introduction of capitation, diagnostic related groups (DRGs) and work on performance related indicators. The pilot testing of these instruments was assigned to subcomponent (d). The fourth subcomponent was dedicated to analyzing and then developing instruments to reduce evasion and improve revenue generation. Within this context, the Government agreed to gradually eliminate the outstanding debt to the CCSS, to continue supporting the cost of care for the indigent and to introduce improved resource mobilization. Component IV. National Health Surveillance and Quality Control Laboratory (US$1.2 million or 3.8 percent of total project costs). Under this component the project was to support the strengthening of the Costa Rican Institute of Scientific Research and Teaching of Nutrition and Health Sciences (INCIENSA) in order to develop a National Health Surveillance and Quality Control Laboratory. This component was included to support the transformation of the MOH into a regulatory body. The component was to finance civil works for the laboratory, equipment and technical assistance. Component V. Project Coordination Unit (US$2.4 million or 7.5 percent of total project costs). The financing for the Project Coordination Unit (PCU) was allocated to a separate project component. This component was to support the technical and financial management of the project. The PCU was to have a project director and unit coordinators in charge of institutional development, administration and financial management. The resources under the PCU were to finance technical assistance, study tours, training, computer equipment and general operational expenditures for project administration. 3.4 Revised Components: There were no formally revised components. While there was no formal revision to the technical content of the components, reallocation of funds were approved in 1996 and 2001, as shown in Annex 2.b, to reflect actual expenditure patterns. 1. CCSS Institutional Reform and Development $ 4,300,000.00 S 2. Redefined Primary Health Care Model, Quality Assurance and Hospital Study $11,100,000.00 S 3. Resource Management and Pilot Testing of Alternative Models of Health Care Financing $ 1,500,000.00 S - 5 - 4. Administration $ 1,900,000.00 S 5. National Health Surveillance and Quality Control Laboratory $ 900,000.00 S 3.5 Quality at Entry: The project design predates the existence of the Quality Assurance Group (QAG) and thus there is no official assessment of the project's quality at entry. However, the preparation of the Staff Appraisal Report (SAR) included a number of aspects related to quality assurance. The ICR finds that the quality at entry process was satisfactory, producing a project that was consistent with national objectives and the Bank's social sector strategy for Costa Rica. The goals established in the Staff Appraisal Report, however, are considered to be highly ambitious for a health sector reform and several of the subcomponents include activities that are not entirely consistent with the overall objective of each component. Considerable support was provided to project design through a Japanese Grant that financed many of the key studies and the development of terms of reference that were critical for project implementation. Substantial documentation, related to key project elements, was provided to the Bank prior to project effectiveness. In the case of the PHC model, for example, the Japanese Grant provided financing for all the detailed design and implementation instruments that were required for a rapid project launch. In addition, CCSS staff actively participated in the preparation of the project, which substantiated Government ownership of the project. While the appraisal team agreed on an extensive list indicators for monitoring and evaluation of the project outcomes, focused mainly on the PHC model, many project activities included in the project lack a performance indicator. The absence of a monitoring and evaluation system, and baseline data, is highlighted as an important limitation regarding the project's quality at entry. Of the 30 key activities that were identified in the SAR, only 14 could be assigned evaluation indicators. Two main subcomponents and component 4 showed no performance indicators. Most of the project indicators focus on impact measures regarding improved health outcomes, coverage, quality and efficiency of service delivery, all of which were part of the overall goals of the project. The indicators that were established to monitor the project outputs, on a component-by-component basis, were focused mainly on the first phase of the project and no further indicators were established. This finding points to the need to establish yearly monitoring indicators on a component-by-component basis. Despite the lack of monitoring and evaluation indicators, the CCSS and the implementing agency kept excellent information on the progress and impact of the reforms. The access to this information facilitated the preparation of a full analysis of costs and benefits (see Annex 3) and enriched the information available to the ICR team. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: The project has demonstrated an important impact on the achievement of the project's intended objectives. Measured in terms of the achievement of overall project outcomes, the result has been highly satisfactory. Table A1.b (Annex 1) outlines the main objectives, proposed actions in the SAR and the key outcomes for each component and shows the high level of compliance with each component. The reform process in Costa Rica made considerable progress in terms of the overall objective to support the Government's efforts to implement critical policy, institutional and - 6 - operational reforms that aimed at improving efficiency, effectiveness and quality of the delivery of health services by the Caja Costarricense de Seguro Social (CCSS). The success of the reform process has been well documented by external evaluations, including PAHO, The National Comptroller's Office (Controlaria General de la Republica) and the PNUD's State of the Nation report. Without the project, it is highly likely that the Costa Rican reform process would have moved slower, suffered increasing delays as administrations changed and reconsidered the reform's direction, and made fewer changes in health financing. The challenges that were not met by the project reflect ambitious initial goals, rather than a lack of political will or technical knowledge to carry out the reforms. The project's achievements can be divided into health outcomes and institutional changes. In terms of health outcomes the project has made important contributions in four areas. First, project interventions directly contributed to extending the primary care model to nearly 100 percent of the population. The official estimates indicate that the total population covered by the redesigned primary health care model reached 88 percent in 2002, with 100 percent of the most marginalized population covered by the PHC model. Despite the official rates of coverage, no resident of the country is denied free access to health care under the Costa Rican system. The official rates show the actual number of people, their dependents and state insured members that are registered with the CCSS. Second, the improved capacity of the primary care services also had a direct impact on the need for hospitalization, reducing the demand in at least seven key areas of avoidable morbidity measured in a recent economic evaluation of the project: acute respiratory infections; diarrhea; prenatal controls; hypertension; intestinal parasitosis; dermatitis; and anemia. In total, an estimated 7,800 discharges were avoided from 1997 to 2001 based on the improvements in primary care, with the principal avoidance in terms of hospitalized cases for diarrhea, prenatal control and anemia. Third, infant mortality was reduced from 14 to 10.8 per 1,000 live deaths over the life of the project. A recent econometric analysis by Luis Rosero ("Tendencias y Perspectivas Demográficas de Costa Rica". Luis Rosero Bixby. Centro Centroamericano de Población. Enero 2003) shows that the reforms in primary care were directly responsible for the reduction in infant mortality by comparing those areas where the reforms were implemented earlier with those areas that had not obtained access to the reformed PHC model. Fourth, hospital infections were reduced significantly through improved monitoring and evaluation under the management agreements and the development of quality assurance mechanisms. Between 1995 and 2002 more than 485,000 hospital days were avoided by reducing the hospital infection rate from over 15 percent to just under 6 percent. The estimated savings from this reduction amount to an average of nearly US$8 million per year or nearly US$47 million between 1996 and 2002. In reduced antibiotic consumption alone, the reduction is estimated to have saved nearly US$4 million over the life of the project. With regard to institutional changes, the project also demonstrates important progress in terms of the introduction of the PHC model, the separation of functions, changes in financial resource management, and the introduction of performance based payment systems. In terms of project outcomes, five main changes are identifiable: First, the introduction of a redesigned PHC model exceeded the project's initial expectations both in terms of scope and scale, reflecting the government's continued political will and technical - 7 - efforts to reform PHC. The project divided the country into 98 health areas (versus the 74 that were originally planned) and established 818 Basic Health Teams (Equipos Basicos de Atencion Integral de la Salud, ( EBAIS)) (in comparison to original estimates of 700), providing for 96 percent coverage, as a share of the total EBAIS required for the entire country, of the redesigned health care model. The investment in equipment, training and materials to support to the PHC model exceeded the proposed component costs of US$16 million. In equipment alone, total project costs in this component surpassed the original estimate by more than US$2 million. In addition, the incremental salaries of MOH staff that were transferred to the CCSS totals more than US$36 million, of which only US$5.8 million was covered by the project. In total, the CCSS invested more than US$30 million of it own resources, in addition to project costs, to support the implementation of the PHC model. Second, the main objective of separating organizational functions within the CCSS was largely achieved. The ICR finds that the original goals were highly ambitious and did not fully consider the link with the complex decentralization process. While some of the organizational restructuring that was envisaged in the project design was not fully implemented, due to political difficulties in managing the change process and technical and timing issues related to the decentralization process, key changes are notable. Following project effectiveness, all pension management was separated from the health and disability insurance of the social security institute. A separate pensions vice presidency was created and full financial separation was established. As a result of the project, cross-subsidization from pensions to health was eliminated and administrative costs were more effectively allocated. In addition, the classical structure of health and maternity insurance Seguro de Enfermedad y Maternidad was replaced with a Seguro de Salud (National Health Insurance) which underscores the universal nature of the CCSS health insurance component. Following CCSS reorganization, steps were made to further separate the CCSS financing and provision. The more notable changes include: (a) the introduction of 133 management contracts (compromisos de gestión) between the CCSS and its 29 hospitals and 98 Health Areas (Areas de Salud), (b) the approval of a Law on Decentralization (Desconcentración), which facilitates the gradual introduction of greater autonomy for health providers, formalizes the management agreements as the governing instrument to regulate the provider network and established 124 health boards (Juntas de Salud) to monitor the delivery of services in all health areas and for all hospitals; and (c) the decentralization process reduced the percentage of administrative staff from over 25 percent in 1994 to 22 percent in 2001. While none of these measures were specifically identified in the project SAR, with these changes the CCSS demonstrated the highest level of political commitment and technical knowledge in the management of the CCSS reorganization and the introduction of the separation of functions. The law on decentralization and its regulations made important strides to improve the incentives facing hospital and clinic directors by removing the "named for life" characteristic of being a hospital director and replacing it with a flexible policies that allow directors to be removed for lack of performance. Under the new law, directors are named for five years and can be reelected or discharged from their position in cases of non-compliance with management contracts. - 8 - Furthermore, it establishes the possibility of hospital and clinic audits by the CCSS Central Administration if they do not adhere to guidelines and management contracts. The Decentralization Law gives hospitals and clinics the power to procure goods and services, including direct purchase of medicines in the market when required. They are authorized to do internal budget modifications and to negotiate external modifications, as long as they adhere to the economic limits set in the management contracts. The efficiency of the decentralization process will depend on the development of better internal managerial capability, a situation that is less serious at hospital level. Third, the introduction of performance-based payments for hospitals, whereby the hospitals receive notional budgets based on the production of services rather than historical budgets, had several important results. The new budgetary system reversed the stagnation in production that was evident prior to the reforms, increasing total hospital production from 295,000 discharges in 1994 to around 340,000 in 2002, while the percentage of relative spending allocated to hospitals fell by around 3 percent. The payment system, which promoted efficiency in terms of the average length of stay, reduced hospital stays from 6.1 in 1993 to 5.4 in 2001. The economic analysis shows that rate of reduction increased significantly with the introduction of the hospital payment systems and the use of diagnostic-related groups (DRGs) as an instrument for clinical management. Greater efficiency is also notable regarding human resource spending. From 1995 to 2001, the percentage of expenditure allocated to personnel fell from over 65 percent to just 59 percent. Finally, new payment systems included a capitated system for primary care. Under this system, all health areas receive a capitated fee. Over the seven years of the project, the introduction of capitation has gradually reduced the inequities in the allocation of resources amongst regions. Fourth, the introduction of a national, fully-automated system for social security contribution collections had an important impact on the institution's revenue efficiency. Achievements are notable in two areas. First, the real revenues per insured person increased by 29 percent between 1996 and 2002, reflecting underlying annual real growth rates of around 5 percent per year, in comparison with pre-reform average annual real growth rates of 2.5%, demonstrating that the project was effective at improving revenue efficiency. The incremental revenues over the 1996-2002 period are estimated at US$175 million. At the same time, the internal efficiency of the system improved. At present 35 of the 72 branch offices are online with the system, providing fully automated revenue collection and improving the billing control in a centralized system. The average billing cycle has been reduced by an estimated 10 days, implying gains of $6 million per year. The system also serves as the basis for the collection of the pension system contributions. CCSS recovers a 0.75 percent commission on pension contributions, thereby making the system self-financing and fully sustainable. And, finally, the automation has reduced congestion at the central offices by improving the capacity of branch offices to process payrolls. Furthermore, many of the large employers now process their payroll contributions online. The government also reduced its net debt burden to CCSS health insurance by over US$12 million, from US$31 million to US$19 million between 1997 and 2002 (Annex 8). - 9 - Fifth, in general value for money appears to have improved significantly with the initial introduction of the contracting mechanisms and the performance-based payment system. CCSS expenditures were reduced from 4.5% of GDP in 1996 to 4.1% in 1999 and 4.3% in 2000 (see Annex 8). At the same time, total public spending, including MOH and other sector institutions, was reduced from over 6 percent of GDP in 1995 to around 5.3 percent in 2001. The total number of discharges, outpatient visits and important health status indicators, such as infant mortality and life expectancy, improved during this time period. At the same time, the strengthening of PHC was accompanied by an internal redistribution of funds from hospitals to primary care. Over the life of the project, the percentage of CCSS health spending allocated to PHC increased from around 15 percent (prior to reforms) to over 23 percent in 2002. This increase was made possible through a relative reduction in hospital spending and concerted efforts to use annual spending increases to disproportionately favor primary care. In summary, the reforms provided important instruments to allow the Government to do more with less. Given that containing escalating health care costs was one of the key reform objectives, it would appear that the CCSS made important progress toward this goal. 4.2 Outputs by components: Component 1: CCSS Institutional Reform and Development Satisfactory. The main objective of separating the organizational functions of the CCSS (policy level, pensions, health insurance and health service delivery) was achieved. However, the two sub-components that-- were not directly related to the separation of functions--Human Resource Development and Pharmaceutical and Medical Supplies System--experienced limited progress. Achievement of the main objective of separating organizational functions within the CCSS is underscored by the separation of all pension management from the health and disability insurance of the social security institute, the 1998 approval of the Law on Decentralization and the implementation of various technical instruments to separate the financing and provision of health care services in the CCSS. A separate pensions vice presidency was created and full financial separation was established. As a result of the project, cross-subsidization from the pensions vice presidency to the health vice presidency was eliminated and administrative costs were more effectively allocated. To promote the separation of purchasing and provision of services, the Health Services Purchasing Department (Direccion de compra) was formally established and staffed to manage the contracting and evaluation process. The separation of functions between the CCSS and the MoH, which aimed at separating the regulatory and stewardship function (to be developed in the MOH) from the financing and provision function (assumed by the CCSS), was also effectively accomplished, thus eliminating the costly duplicity in personnel and infrastructure. During the process, around 1600 staff members were transferred from the MoH to the CCSS. However, the full development of the MOH as a regulatory and policy-making body has lagged behind. Lack of technical capability and political support, and as a consequence insufficient budgetary allocations, appear to be the main causes. Despite the progress in separating the CCSS main functional areas, the internal reorganization of - 10 - the CCSS lags behind the development of instruments and a legal framework to support the decentralization process. Interviewers unanimously identified the lack of political will and leadership to control the interest groups (feudos de poder) within the CCSS as the main obstacles to achieve the separation of functions and decentralization. The CCSS Board of Directors approved the separation and decentralization of functions in 1997, so that the directors of the period 1998-2002 were faced with the implementation of the agreement. While many of the technical complications regarding decentralization have been addressed, there are many outstanding issues regarding the distribution of rights and responsibilities between the central and local levels. The delays in implementing the CCSS reorganization are often cited as a key hurdle for the reform process. There has been steady progress in implementing the decentralization of functions proposed by the project. The Law on Decentralization was approved in 1998, followed by the approval of the regulations governing the law, which were approved in 1999, and implementation initiated in 2001 with 15 health care providers. The decentralization has been gradually implemented with a total of 34 centers (out of 133) operating under decentralized rules by 2002. These 34 centers, however, constitute nearly 80 percent of the total CCSS health spending. The project supported the creation of Health Boards (Juntas de Salud) and strengthened planning processes at the CCSS with the introduction of a service delivery plan "Plan de Atención a la Salud de las Personas" as an instrument to link health resource investment with community health needs. The project also supported the development of a semi-autonomous superintendency within the CCSS to promote quality assurance and users' rights. The Superintendency, denominated SUGESS, provides users with a network of health services ombudsmen to voice concerns over treatment and quality, evaluates the quality of services and provides an unbiased evaluation of management agreement targets. The SUGESS was an initiative that developed internally, without prompting by the project framework, to ensure transparency in the separation of functions and to increase user's rights within the system. While the SUGESS could have been given more autonomy and authority as an external entity to the CCSS, a strategic decision was made to introduce these functions within the CCSS organizational structure and to later consider greater autonomy. The project trained 232 administrative staff and 14,888 course and seminar participants. Roughly 60 percent of all CCSS management received training under the project. The project also supported the development of five masters programs: (a) Masters in Health Economics (66 graduates); (b) training in Public Health Management (36 graduates); (c) Masters in Epidemiology (16 graduates); (d) Masters in Public Health (15 graduates); and (e) Masters in Health Services Administration (18 graduates). In addition to these programs directly related to developing management capacity, hundreds of personnel were trained in change management, specialized postgraduate programs in accounting, nursing, PHC model and public administration. Overall interview responses show that training quality was high and adequate in quantity and timing to respond to the organizational changes targeted by the project. The project maintained a concerted effort to include communication efforts throughout the reform process. The PCU employed a press officer and numerous activities, materials and supplies and consultancies were financed under this component to inform the internal and external clients of the reforms. The communication efforts were critical to the project's success. - 11 - The SAR projected an expenditure of $1.4 million in order to accomplish the separation of functions and decentralization in the health sector. The actual spending, however, represented only 52% of this amount or $729,000, as most of the activities were carried out under component III. Besides the separation of functions and decentralization, component I included the improvement of the manpower planning and the drugs and medical supplies system, and the achievement of a more rational use of medicines. The former activity included studies to analyze existing norms and staffing, personnel incentives and civil service norms applied to the central and regional offices of all three CCSS units. The project supported an extensive study by the University of Costa Rica regarding alternative staff employment arrangements, however, the recommendations were not implemented. At the same time, improvements were not achieved in the CCSS-owned pharmaceutical and medical supplies system. Additionally options were to have been explored for private contractors to manage the handling and distribution of supplies. Little progress was made to improve the rational use of medicines by introducing cost-effective prescribing, correct dosage and improved communication to patients. The component envisaged the development of a fully-integrated information system to address all of the CCSS management information system needs. The project supported the development of an information system for the purchase and inventory management of drugs and supplies. The overall MIS was not completed for several important reasons: (a) the resources required to finance a project of this nature were not included in the project; (b) the delays in decentralization made it unclear how functions would be distributed amongst the levels; and (c) key areas were addressed initially, such as the billing and pharmaceuticals, and the overall approach was left for future development. In conclusion, the main objective was achieved and the project was key to the reorganization of the CCSS. The delays or limitations in achieving some of the additional objectives are mainly due to the difficulty of organizational change and the feasibility of implementing institution-wide projects with limited resources. Not surprisingly, the component's total expenditure of $3.4 million is only 35% of the projected amount in the SAR of $9.8 million. A straightforward evaluation of this component, however, is complicated by two compounding factors. First, the implementation of many of the activities proposed under this component was carried out under Component III, Resource Allocation. The CCSS made a strategic decision to combine the studies and implementation of reforms regarding the separation of functions and the introduction of performance-based payment mechanisms. As a result, many of the disbursement targets and outputs are allocated to Component III. Second, although several indicators were defined in the SAR to monitor the component's activities, output data are limited mainly to training. Component 2: Redefined Primary Health Care Model, Quality Assurance System, and Hospital Study Highly satisfactory. The overall component objective of implementing the redefined PHC model in the CCSS, including the transfer of functions from the MOH to the CCSS, was fully accomplished. The redefined Primary Health Care Model (PHC model) is probably the most significant achievement of the reform process supported by the project. Coverage was assured by creating 102 Health Areas in the country's territory (98 are already installed or 96%), each of which comprises around ten to fourteen basic medical units called EBAIS. In total 818 EBAIS have been established covering 88% of the country's population. - 12 - Medical equipment was acquired and new personnel were hired to implement the new PHC model. The model was fully implemented in the poorest areas of the country. In order to introduce the integrated health care packages for children, adolescents, adults, elderly and women, 222 health professionals and 1274 auxiliary staff were trained. The transfer of around 1600 staff members from the MoH to the CCSS not only underscored the separation of functions between the Ministry and the CCSS, but also made the successful implementation of the PHC model possible. The process was accompanied by a shift of budgetary allocations toward the primary health care level, increasing its share of health insurance expenditure from 19% in 1997 to 22% in 2002. This component played an instrumental role in achieving the project's overall objective of extending access, increasing efficiency and improving quality. The extensive efforts to implement the PHC-model produced an expenditure of $48.8 million, representing 305% of the projected spending in the SAR of $16 million. The main element contributing to this overspending was the incremental salary of new staff of $36.6 million, thus by itself exceeding the project's total budget of $32 million and the allocation for incremental salaries of US$5.8 million. In part, the difference between the initial estimate for incremental salaries and the actual project expenditures is due to the additional 2 years of project implementation, the increase in the total number of EBAIS that were implemented to ensure adequate population coverage and additional salary increases that were negotiated during the project period. Nonetheless, the original design of the project did not adequately estimate the incremental cost of the model. During the implementation process, CCSS has identified several issues that should be addressed in the future. While these aspects were not necessarily part of the original project design, they will be important for the effectiveness and sustainability of the model. As such, the Bank-financed Second Health Reform Project that is already under execution will address many of these issues. There are four main outstanding issues: First, despite the incorporation of quality indicators into the management agreements and the development of protocols for primary care, the deployment of a full-scale continuous quality improvement model has not been implemented. The transition has been slow to a new care paradigm, with integrated health services, including family doctors, community services, promotional activities and the emphasis on preventive health services as a mean to improve cost-effectiveness. While important advances have been made in the development of protocols, progress is limited in implementing effective referral pathways between primary and secondary and between primary/secondary and tertiary level facilities. There is a lack of coordination between the primary level and the secondary and tertiary health service levels. Second, the rapid deployment of the PHC model has outpaced the development of managerial capacity in the health areas, including some limitations regarding equipment at several EBAIS and slow development of an information system to monitor and evaluate PHC services. In reality, the political pace of implementation appears to have outpaced the technical pace of implementation, creating some gaps in the model. Third, the development of Health Boards was an excellent initiative to increase community participation. Now support from community organizations in defining the health services mix and in participating with quality control needs to be improved. Training quality also needs to be - 13 - improved for the members of the "Juntas de Salud". Appropriate incentives for community members to participate in decision making should be incorporated into the model. Fourth, significant progress was made towards increasing competition in the provision of health services but not as much as was originally envisaged in the SAR. Regulations were approved by the CCSS Board to govern the contracting of private providers, management contracts were designed and implemented to regulate the private provision of care with CCSS funds, and important progress was made in developing the capacity to evaluate the private provision of care under these contracts. The CCSS increased outsourcing of ancillary services, contracted out nearly 30 percent of the population's under the primary care model to private and cooperative providers in metropolitan area, and it increased the purchase of diagnostic services from private providers. However, little progress was made in introducing competition at the hospital and secondary care level. Component 3: Resource Management and Pilot Testing of Alternative Health Care Financing Highly Satisfactory. By fully implementing changes in resource mobilization and resource allocation, this component surpassed the component's objectives which were focused on developing studies and pilot projects to improve resource mobilization and resource allocation. This subcomponent expenditure's reached $4.29 million, representing 477% of the spending projected in the SAR of $0.9 million, reflecting the execution of many of the activities originally designed as part of the Component I. The project made important progress in introducing mechanisms to promote greater equity and efficiency. Improvements in the equity of resource allocation are evident at the primary care level, with the capitated payment mechanisms implemented in 1997. Over the past 5 years, the variation in per capita expenditures among Health Areas has been significantly reduced. Improvements in efficiency are notable both in terms of the shift on resource allocation towards primary care with relative percentage of funds received by PHC from around 15 pre-reform to around 23 percent post-reform. The average length of stay at hospitals decreased and production of discharges and consultations increased more than 10 percent and 32 percent, respectively over the 1996-2002 period. Many of these gains were made possible by the introduction of management agreements, entitled Compromisos de Gestion, between the Health Services Contracting Department and the hospitals and health area providers. The management agreements clarified the results expected from providers and introduced for the first time indicators to evaluate the performance of providers with a link to resource transfers. This contributed to greater transparency and to an "efficiency culture" that never existed before. The component introduced for all 29 hospitals a performance-based payment system, based on the introduction of a unique unit of hospital production--the UPH (unidad de produccion hospitalaria). This production unit allowed the CCSS to link payments to production and replace the historical budget for hospitals with a performance-based budget. The system was implemented and tested at first as a shadow-payment mechanism and then in 2000 linked to the budget. Even before it was actually linked to the budget, the payment system served to stimulate hospital production and the number of surgeries, reduce the average length of stay, and create a culture of performance. - 14 - In practice, it has been difficult to punish providers for lack of compliance and incentive systems have not yet been implemented to reward individual employees for stellar performance, although they have been developed and approved by the CCSS Board for implementation. As some interviewees commented, the main reason for the delay in fully linking resource transfers to performance indicators and the achievement of goals is that, in the end, any sanction imposed on a public health provider will be paid either by the CCSS itself or by the insured in the form of a even more severe rationing of health services, but not by those individuals responsible for the failure to meet contract targets. To this extent, the implementation of the payment mechanisms cannot be considered independently of the implementation of decentralization. While the instruments were fully developed and deployed under the project, there are number of political aspects, critical to the long-term success of the model, that have not been addressed. In particular, sanctions affect the provider's budget, but not staff salaries, incentive payments or employment conditions; and providers face a captured demand which limits incentives to increase efficiency and quality. The CCSS has designed additional actions to increase competition at the hospital level but these measures have not been implemented. The second subcomponent targeted health insurance revenues. It was intended to reduce evasion and increase affiliation, to analyze the feasibility of co-payments, deductibles and cost-recovery of services to non-insured, to seek to reduce the outstanding Government debt with the CCSS, and to evaluate the financial effect of alternative delivery systems such as "medicina de empresa", "medicina mixta" and "libre elección". The main objectives of this subcomponent were fully achieved. Driven by the increased expenditure on information systems, total expenditure of this subcomponent amounted to $4.37 million or 1456% of the 0.3 million projected in the SAR. A major effort was undertaken to improve the collection of insurance contributions and to better train staff members. The most notable improvement was the development and deployment of the universal revenue collection mechanism, Sistema Centralizado de Recaudación (SICERE) which was instituted in 2000 to centralize all health and pension premium collections. The new system has been deployed in more than half of all CCSS branch offices resulting in estimated savings of US$6 million per year due to reduced processing times.. In 2000, the Law on Worker Protection (Ley de Proteccion del Trabajador) was approved marking the introduction of broader powers for the CCSS' efforts to reduce evasion. The Law also promotes the universalization of contributions introducing mandatory contributions for all formal and informal sector workers. Despite the introduction of the Law, little progress has been made in increasing total affiliation. As a percentage of the total economically active population, the share of insured remains stable at around 65 percent. The driving factor behind CCSS affiliation, however, has to do with wider macroeconomic and labor market conditions--several years of recession have cut deeply into informal employment, reducing the effectiveness of simple comparisons of formal insurance. Major reductions were made in Government debt to the CCSS both as an employer and as contributor for the indigent. Between 1993 and 2001, the average debt (estimated as the difference between what the Government should pay and what is actually sent to the CCSS) was reduced from over 65 percent to just over 14 percent (See Annex 8). In real terms, the accumulated debt between the Government and the CCSS was reduced from US$31 million to - 15 - US$19 million, representing a net revenue gain of US$12 million for the Health Insurance. This has had a major impact on CCSS revenues and was an important by product of the project negotiation which brought light to the magnitude of the debt and established targets for debt reduction. At the same time, the outstanding debt to the CCSS was also reduced significantly, with the loan obligations assumed by the Ministry of Finance (MOF). The project also managed to promote the interchange of information between the CCSS, the MOF and the National Insurance Institute (INS) which triangulates reported income between the three institutions and thereby allows inspectors to more effectively target evasion. Another major goal of component III was the development and subsequent testing of Diagnostic Related Groups (DRG) that facilitated clinical management. DRGs were successfully introduced in 24 out of 29 hospitals and 850 health staff were trained in the use of DRGs. Interviewees recognize the usefulness of the DRGs to improve inpatients´ information, support health management and reduce costs in hospitals. The DRG implementation in Costa Rica was the first attempt to implement the full DRG model, with 511 groups of procedures, in Latin America. The process resulted in a significant learning experience. Some of the key lessons learned include: (1) The introduction of DRGs is a complex task that involves high consensus and a great amount of training. Although the DRGs were originally envisioned to serve promptly as payment mechanisms, its introduction has been quite successful as a clinical management tool oriented to benchmarking hospital performance, but has not yet served as a financial reimbursement mechanism; (2) the failure to create financial incentives to use the DRGs limited the interest of directors, administrators and clinicians in the use of the tool; (3) training to use DRGs was insufficient and many users continue to perceive them as complicated, almost not understandable, and difficult to use. In fact, DRGs are not used in the training of senior and middle management; (4) it is perceived that DRGs were not adapted to the "Costa Rican health model" or the "Costa Rican hospital culture", as several interviewees pointed out; (5) the need to allocate sufficient resources to renewing licenses and providing ongoing technical support and training to staff were not fully estimated; and (6) DRGs were not integrated into the hospital information system and therefore they are not regarded as part of an integrated management system. As an alternative, gradual phasing in could have been reached by defining a slowly increasing fraction of budgetary assignments subject to results in terms of DRGs. In addition to the introduction and testing of DRGs, this subcomponent also included the review and pilot-testing of fixed capitation transfers, improvements in hospital administration supported by the development and introduction of the MIS-module in hospitals (Management Information System on hospital administration), testing of the referral system and the increased participation of private providers. In these areas, progress is notable in two of the four areas. Capitation payments were introduced as part of the separation of function under the management agreements. At present, all health areas receive their annual budget based on the total population, without risk adjustment, eliminating the historical budgeting that characterizes the pre-reform phase. At the same time, private providers and NGOs assumed an increasing role in the provision of primary care services but were not financed by the project. During the project period, management contracts were extended using NGOs and cooperatives to cover more than 300,000 people in the metropolitan area. These contracts have been established with the University of - 16 - Costa Rica, cooperatives and a private, non-profit provider. The provision of services under these contracts has shown excellent results in comparison with the traditional CCSS model, indicating that the private sector can provide the PHC package at a considerable cost-saving in comparison to the traditional, public model. Expenditure for the DRG systems and resource allocation technical assistance amounted to $1.37 million or 228% of the $0.6 million projected in the SAR. For pilot-testing several alternative delivery and financing systems only $0.086 million or 17% of the projected $0.5 million were used, however actual expenditures, assumed by the CCSS, amount to more than US$10 million per year. Component 4: National Health Surveillance and Quality Control Laboratory Satisfactory. Component 4 achieved its intended objective to strengthen the Costa Rican Institute of Scientific Research and Teaching of Nutrition and Health Sciences (INCIENSA) by establishing a National Health Surveillance and Quality Control Laboratory. The laboratory facilities for infectious diseases and febrile and hemorrhagic diseases were rehabilitated, contributing to Costa Rica's ability to detect and control TB and strengthening INCIENSA's capacity to provide overall epidemiological surveillance in these areas. An integrated information system was developed and implemented facilitating INCIENSA's management of critical information. INCIENSA staff received considerable training and technical assistance. This component was executed entirely by INCIENSA and was dependent on the MOH budget for counterpart funds. This organizational separation led to some problems in coordination with the project's PCU. Furthermore, limited MOH counterpart funding slowed down the component's execution. Component 5: Project Coordination Unit Component V provided ongoing support to the project's administration. The project was successfully managed during the seven years of project implementation, accruing a total expenditure of US$ 2.37 million out of an allocated US$2.4 million. Between 1995 and 1998 the PCU was dependent on the Executive President's office of the CCSS, facilitating direct communication and a high level of execution. Starting in 1998, the Modernization Vice Presidency was created at the CCSS to be in charge of the coordination of all external financing supporting the reform process at the CCSS. The PCU then became a sub-unit of the Modernization Vice Presidency. This change was accompanied by staff changes, and execution slowed considerably while the project adjusted to the new organizational arrangement. To its credit, after a 6-8 month delay, the PCU quickly resumed implementation and overcame the short-term challenge of reorganization. The PCU developed significant financial management capacity although not always at the level required by Bank standards. Although it was not included as part of the original project design, a major deficiency was the failure to implement an adequate monitoring and evaluation system. 4.3 Net Present Value/Economic rate of return: The reform process produced significant efficiency gains and improvements in the population's health that have been estimated to yield benefits with a net present value (NPV) of US$86 million and an economic rate of return (ERR) of 69 percent. The present value of the project's benefits was US$254 million, far exceeding the original SAR estimates of US$41 million in total benefits. - 17 - The cost-benefit analysis that was prepared as part of the ICR process includes benefits in the following areas: (a) reduced hospitalization for avoidable morbidity including: parasitic infections, respiratory infection, diarrhea, hypertension, anemia, pre-natal care and dermatitis; (b) reduced hospital infection rates which decreased the use of antibiotics and bed days; (c) a reduction in the Average Length of Stay, yielding many savings in variable costs; (d) the reduction in infant mortality; and (e) reductions in the process time for the CCSS billing system that yields important savings in financial costs. 4.4 Financial rate of return: N/A 4.5 Institutional development impact: Highly satisfactory. Several positive institutional impacts of the project are clear. The separation of functions between the CCSS and the MoH was effectively accomplished. The CCSS has initiated its restructuring into independent vice presidencies (starting with pensions, health care and financial management). And, the health care unit has introduced a clearly defined Primary Health Care level and 7 regional offices for administrative decentralization. Second, all of the significant project accomplishments-- separation of functions, PHC model, revenue mobilization, resource allocation, DRGs and decentralization-- were fully institutionalized. The implementation and coordination of the project constitutes an invaluable experience for those who participated closely in the process. Consultants were qualified by the Bank, but selected and contracted by the CCSS. They attended evaluation, coordination and follow up meetings to monitor the accomplishment of targets and the execution of budgets. National counterparts had to present follow-up reports monthly or every two months. The regular monitoring and evaluation has been beneficial to both senior management at the central offices and to those in the periphery directly involved in the execution of reform actions. The institutional capability to develop and implement reform programs and to collect and analyze data to assess program progress has been strengthened. The Bank procurement procedures used during the project were regarded by the interviewees as efficient, fast, clear and in accordance with national legal framework. Conducting Bank procurement processes during the reform project has clearly strengthened institutional capabilities and the PCU developed significant skill in managing complex, dual bidding procedures (see 5.2, below). 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: Three changes in government posed important challenges to the project's sustainability but did not alter the course of reform. Throughout the three changes in government, the project's main reform objectives were maintained and the principal activities continued. Nonetheless, each change of government implied at least 12 months of delay. During the first change of administration, in 1998, there was a short period during which the project was rated as "unsatisfactory" as the new administration froze implementation to reconsider the depth and pace of reform. Once the new administration affirmed the positive direction of the reforms, project execution was resumed and the "satisfactory" rating was once again obtained. As the 1998-2002 administration evolved, important measures were taken to consolidate the reform process by approving substantive legislation in support of decentralization, management agreements and the efforts to - 18 - improve revenue mobilization. Overall, the CCSS was quick to adjust to changing situations. 5.2 Factors generally subject to government control: During the period 1996-1998 the project was subject to restrictions in financing from the Ministry of Finance. This reduced the pace of implementation. In addition, the failure to fully implement the procurement process under Bank procurement guidelines, due to ambiguous rulings by the National Comptrollers Office (Contraloría General de la República) produced long procurement processes, as the CCSS was forced to carry out the procurement process under both Bank and National procurement rules. 5.3 Factors generally subject to implementing agency control: During the first four years of the project, no coordinator was assigned for component I, Institutional Strengthening, thereby reducing the effectiveness in the implementation of this component (in the original project design, the PCU was not assigned a coordinator for this area). Lack of clear definition and an organizational mandate in this area reduced execution in this areas. Overall during the life of the project, most of the PCU staff was highly stable, producing a high level of continuity and increasing specialization. However, the failure to introduce new personnel reduced the overall effectiveness of the PCU and contributed to organizational inertia. 5.4 Costs and financing: Total expenditures incurred for the project, including consulting services, works and equipment financed by CCSS funds were US$65.4 million, or 204% of the total estimated project cost estimated at appraisal of US$32 million. Disbursements reached 98 percent of total project funds. Most of the cost increases were incurred in the additional equipment required for the redesigned PHC model and the incremental salaries for staff transferred from the MOH to the CCSS or employed to increase the capacity of the PHC network. The disbursement cycle suffered delays, principally in the years prior to and post elections (1998 and 2002), extending the project cycle from 5 years to over 7 years. The total project cost of $32 million was to be financed by the Government and the Bank at 34% to 66% ratio, respectively. However, the Government ended up financing more than 67% and the Bank at 33% of the actual total expenditures. The most prominent procurement method envisioned at appraisal was other methods (56%) followed by international competitive bidding (ICB: 17% of total costs), local competitive bidding (NCB: 9%) and non-bank financed (NBF) at 18 percent. In reality, procurement methods followed closely the original allocation in the SAR, with the exception of the dramatic increase in NBF due to the CCSS incremental contributions to the project. In sum, ICB was used for 10% of the total cost, while LCB was used for only 6% of the total costs, NBF was the most common procurement method with 56 percent, and other methods with 29 percent. 6. Sustainability 6.1 Rationale for sustainability rating: Highly Likely. The reform measures introduced under the project are very likely to be sustained. Under component I, the actions that were taken have been fully institutionalized, including the approval of a Decentralization Law and a full regulatory framework for its implementation. In addition, the separation of functions has been fully integrated into the institutional framework. The training provided has been integrated into the CCSS training unit, CENDEISS, and courses developed under the project are now part of the standard curricula for health professionals. - 19 - The actions under component II are now fully integrated into the CCSS, as the last 3 years of the project witnessed full control of the implementation process by the CCSS. Moreover, steady increases in the share of funds to PHC (rising to 23 percent in 2002) provide for a sustainable basis of the investment made under this component. Component III has also made important progress in institutionalizing the changes. All changes in resource allocation mechanisms have been assumed as a part of the standard budgeting process, the management agreements are now signed with all CCSS providers, public and private, and fully govern the objectives and payment mechanisms for providers. The SICERE and changes in revenue generation are fully sustainable as both have independent legal frameworks (Ley de proteccion al trabajador) and are self-financing. The areas where additional support will be required to fully institutionalize the instruments developed under the project include: full implementation of the DRG system and the more profound changes in payment mechanisms to hospitals, shifting from a budget based on production to a revenue based budget based on the services produced and placing hospitals at risk for compliance with the management agreements and the performance targets. The actions under component IV are expected to be sustained as they have been fully assumed by INCIENSA. The Government prepared, with the Bank's assistance, a second project that addresses many of the issues that are pending consolidation from the first project, such as reorganization, financing mechanisms, DRGs and others. The Second Health Reform Project is already effective. For the short- to medium-term, policy changes intend to: (a) align the organizational and functional structure of the CCSS with recent changes separating financing, purchasing and provision of health services at all levels; (b) promote improvements in quality and fulfillment of consumer rights, as well as the efficiency and effectiveness of the Ministry of Health (MOH) as a regulatory agency, by strengthening the institutional and regulatory framework; (c) improve the quality and efficiency of the CCSS health delivery system by supporting decentralization of decision-making, consolidation of the primary care delivery network based on a population-based system, and introduction of performance-based incentives for providers; (d) reduce inefficiencies in the pharmaceutical sub-sector and promote rational drug use by introducing changes in the planning, purchasing and distribution of pharmaceuticals and supplies; and (e) develop financial mechanisms that will improve the equitable distribution of resources, improve efficiency in the provider payment mechanisms and strengthen the CCSS capacity to collect payroll contributions. 6.2 Transition arrangement to regular operations: As cited in 6.1, most of the activities financed by the project have been fully integrated into the CCSS organizational and functional structure. The CCSS vice presidencies maintain full control over the instruments developed and continue to support their full implementation. Furthermore, the PCU has been mostly maintained for the execution of the Second Health Reform Project that was approved on June, 2001. 7. Bank and Borrower Performance Bank 7.1 Lending: Satisfactory. The participation of the Bank in the design and consensus building related to the reforms was considered exemplary by the key stakeholders interviewed during the ICR preparation process. The Bank contributed to the policy debate during the design phase and provided important support to the transition period between governments. The CCSS had four different Executive Presidents since project design started in 1992. Three Government administrations carried out the project over time and a fourth one participated of the last six months of project life. During this period the Bank served as a catalyst and - 20 - helped to build consensus for the reform process among different political groups. The main deficiency is related to the inadequate estimation of the recurrent cost of the PHC model and the underestimation of the cost to implement complex organizational reforms and information systems, for which sufficient resources were not provided under the loan. The initial project design also established highly ambitious goals for the CCSS, underestimating the complexity of the reform process. 7.2 Supervision: Highly Satisfactory. In general, Bank supervision efforts were satisfactory. The project had four Task Managers over the life of the project but the PCU and the Bank maintained a common line and sophisticated policy dialogue. Two Bank missions a year were standard and close communication by video conferencing helped to maintain a close relationship with the client. On particular occasions the client was invited to Bank Headquarters to discuss particular issues and in general to ease the transition between one government administration to the next. Lack of more concise and easily accessible project indicators from the early stages of the project made it difficult to evaluate progress with concrete measures. However, supervision efforts used client reports and evaluations of the reform process as a guide for monitoring project progress. In general the Bank was flexible and accommodated new situations and demands without compromising the technical quality of its advice and the progress of the reform. 7.3 Overall Bank performance: Highly Satisfactory. The Bank has been a major supporter of the health reform process in Costa Rica. Through an excellent partnership with Costa Ricans, the Bank has supported the dissemination of the health reform process both internally within Costa Rica as well as internationally. There is growing interest in the Costa Rica experience in the Latin American Region. The Bank demonstrated considerable flexibility with regard to the technical and policy contributions to the reform process and was a valuable partner working to ensure continuity of the reforms. On the administrative side, Bank procedures were often an additional complexity, on top of already cumbersome local procedures, which were an obstacle to efficiency. Another shortfall regarding Bank performance is underscored by the lack of a monitoring and evaluation system and the reformulation of monitoring indicators that were relevant to the project's implementation. During supervision of the project, the Bank emphasized the identification of lessons learned overtime and applied the lessons from the project to the preparation of a new project that continues to support the reform agenda. Borrower 7.4 Preparation: Highly Satisfactory. The government was very active in the preparation of the project. Through activities financed by the CCSS and the Japanese Grant, detailed preparation work was carried out to prepare the CCSS for rapid deployment of the PHC model and other key issues identified during preparation. The borrower also took a proactive approach to building consensus among key stakeholder, prior to project implementation, which allowed the project to rapidly launch key activities. 7.5 Government implementation performance: Satisfactory. Strong government commitment to the project objectives was key for implementation success. The government gave high political visibility to the reform process and maintained its goals as priorities in the political agenda from 1994 to 2002. The Borrower assumed the project in payment of its internal debt to the CCSS. It satisfactorily reduced is overall debt to the CCSS over the life of the project. Evasion in Government contributions to the CCSS as employer and as responsible party of contributions for the indigent was drastically reduced during project life (see Annex 8). Two aspects under government control influenced implementation pace: there were difficulties in the availability of counterpart funding in the case of the INCIENSA and delays in procurement processes were present when applying both national - 21 - and Bank regulations to project implementation. (Section 5.2). The cumbersome national bidding process thatch forced the PCU to follow both national and Bank procurement guidelines slowed the process considerably. In this regard, the Contraloría General de la República did little to facilitate the bidding process. Second, having two implementing agencies complicated project management; especially when one is more dependent on the National Budget than the other. 7.6 Implementing Agency: Highly Satisfactory. The CCSS and the MOH demonstrated high commitment to the reform from the design of the operation through implementation. The PCU developed an excellent capacity to manage the Bank financed project and to promote the institutionalization of the reforms. For a brief period in 1998, as the administration and PCU staff changed with the changing administration, the PCU experienced a considerable slow down in execution and limited capacity to attain the momentum the reforms had achieved from 1996 to 1998. This resulted in an "unsatisfactory" rating. Once the new team understood the reform process and Bank procedures, the reform process resumed its previous rhythm and the project returned to a satisfactory rating. A highlight of the reform process is that most of the key initiatives financed by the project have been fully assumed by the CCSS line departments. This speaks to the work of the PCU in developing excellent coordination with the CCSS departments and in taking an unselfish approach to facilitating the reform's implementation. 7.7 Overall Borrower performance: Highly Satisfactory. The government accepted the challenge of implementing a complex reform process and maintained the reform path over four different administrations. As an addition indication of borrower performance, the CCSS was able to finance with its own resources the consistent implementation of reform initiatives, even as Bank disbursements declined in 1998 and 1999. 8. Lessons Learned This section is divided into three parts. A more general part addressing project design and implementation issues is followed by a discussion of lessons learned organized according to the main strategies of the reform process: (1) overall comments on project design and implementation; (2) issues related to the separation of functions, resource allocation and reimbursement mechanisms (including DRGs); and (3) the redefined primary health care model. While this is not an intensive learning ICR, the preparation process has considered extensive stakeholder analysis based on the results of a qualitative and quantitative evaluation of the project. On the qualitative side, an external consultant was contracted to carry out more than 30 interviews with key stakeholders, organized in five modules These interviews provide an in-depth view into the projects objectives and outcomes. This was complemented with an in-depth review of the project objectives, activities and outcomes, in the context of a logical framework that substantiated many of the ICR findings regarding project implementation. Following approval of the ICR, the Bank will hold a workshop with the CCSS to discuss the ICR findings and the lessons learned. Part One. Project design and implementation 8.1 There are minimum necessary preconditions for a successful health sector reform project. These are: i) a multi-party political agreement in favor of the reform process within a general political context that regards change and reform as a mean for progress; ii) the existence of sound sector studies that have induced in the leading technical team a strong sense of the need for reform and have led to the development of a sound project concept, complemented with external financing; and, iii) a strong communication campaign to build and maintain national consensus. Under these conditions the project creates strong positive expectations in key players at the start of the reform process. Politicians will - 22 - speak out publicly in favor of the reform and the clarity of objectives and targets constitute the best basis for communicating reform benefits to a broader public. External financing gives the feeling of security both because of its implicit technical backing of the reform design and the financial sustainability it is conferring. Additionally, the Bank usually acts as a catalyst to bring together different political actors around a technical proposition and accelerates the process of project building. In general, these conditions are likely to produce a strong positive momentum for reform which is essential at the initial stage. 8.2 A reform project should be understood as one of the instruments supporting changes in the sector, where the achievement of goals and objectives is enabled or impeded by a complex interaction among political, social and economic factors. Project success is strongly dependent on the strategic use of such combination of factors both at design and during implementation. There were key policy decisions that enabled the achievement of project goals: i) the CCSS decided not to carry out pilots and to launch reform processes at the national level instead, in order to reach the "point of no return" rather than risking a pilot project that could be overturned; ii) after creating the information system for the collection of contributions SICERE, the "Ley de Protección al Trabajador" was designed and passed which not only accelerated the implementation of the system, but enabled the CCSS to enhance the benefits offered by accessing to compiled and updated information for evasion reduction; iii) the design of the reform project strategically linked CCSS and MOF interests to support the reform endeavor. With the project not only the MOF was reducing its debt with the CCSS but the latter was receiving the technical support its reform agenda required. 8.3 Reform processes should identify "early wins" that will provide buy in and political capital to make more difficult structural changes down the road. CCSS leadership identified early on one key aspect, the extension of the PHC model, that would be of direct interest and impact to the common person in Costa Rica and would strongly influence public opinion of the policies to be implemented. Visible results on access to health care services and impact on health status could be achieved through an aggressive approach to the implementation of the redefined PHC model in combination with a strong communication campaign at all levels. The strategic decision to start implementation of the PHC model in the poorest cantons and to develop management agreements with those providers that represented more than 55 percent of all expenditures were critical to achieving early success with political commitment. 8.4 Reform processes are costly endeavors and both the Bank and the client should not address timidly the estimation of such costs. To ensure the financial sustainability of a reform, it is necessary to make accurate cost projections for those components that represent a large fraction of total project budget as well as for transitional costs in those areas where the reform process results on, at least at the beginning, a serious cost escalation. Project cost estimation of introducing the redefined primary health care model was $16 million, but the incremental salary of new staff alone required $36.6 million (Section 4.2-component 2). Part of the reason for this cost increase is an underestimation of the required size of the primary health care network capable of responding to an increasing demand for services induced in part from the benefits of the reform process. Estimations at appraisal foresaw 800 EBAIS in 74 Health Areas, or a ratio of 5000 persons per EBAIS, whereas by 2002, 819 EBAIS have been established in 94 Health Areas for a ratio of 4200 persons per EBAIS. In this case the CCSS was fortunate to be in a financial position to take over this unexpected cost increase, but in countries under different circumstances an error of this magnitude may disrupt completely the reform project and create a more costly situation for the sector than that before the project. 8.5 To measure the impact and outcome of each activity of the reform project, it is not enough to define monitoring and performance indicators at appraisal. The project must create a system capable of providing and analyzing such information regularly so it can be evaluated throughout the implementation period. Collecting and analyzing baseline data for this purpose must be part of project design. Of the 30 activities that were identified in the SAR, only 14 could be assigned performance - 23 - indicators and 2 activities had at least one indicator that could be evaluated with available data. Fortunately, there is a lot of information collected in Costa Rica that enabled us to gather results of the reform process from different sources and complement information obtained through the interview process. 8.6 Project design should maintain an internal equilibrium between different beneficiaries of the reform and the timing of benefits. In order to maintain the support for the reform process it is important to avoid great disparities in the benefits among the different groups directly involved in the reform, for example management, health staff, hospital personnel, primary care level personnel, etc. In an analogous way, it is very helpful to maintain reform momentum if the first benefits become evident shortly after the process has begun and continue to be felt more or less evenly distributed over the entire implementation period (see also section 8.3). 8.7 The implementation unit of a project should be adequately staffed and prepared to successfully deal with several political challenges, such as eroding political support. As sector projects usually extend over several government administrations and priorities change over time, internal resistance from groups that lose power and influence, are misinformed or simply oppose change, can result in general apathy towards the project if benefits do not become evident fast enough. There are several strategies that can be followed to overcome these situations. First, the Bank is a key player as it acts as an external coordinator, guiding the process by contributing high quality consultants, creating areas of discussion that are important for the continuing effort of consensus building, defining methodologies and courses of action, and financing usually the larger part of project costs. The loan contract provides continuity to the project, as it obliges changing governments, and with them changing key actors, to comply with contract terms. Additionally, the Bank takes over the supervision and reviews the auditing of the project. The Bank should use this leading role to maintain project schedules and Bank procedures, and although a sensitive aspect, the Bank should enforce strong leadership on the creation of adequate PCUs at all times. The potential of the Bank to guide the general reform course must be exploited in a consistent and permanent manner respecting a delicate equilibrium between national ownership of the process and Bank's guidance. Third, constant communication is critical during the reform process. It reduces internal resistance as targets become clear and effects foreseeable. Fourth, strict timetables to implement each activity of the reform and the complementary information campaign can effectively reduce uncertainty and combat general apathy that inevitably arises as reform is implemented. Fifth, consensus building between key players from the civil society, sector labor unions and politicians is a continuing process that should be carefully attended at all stages of the reform process. 8.8 Depending on the specific characteristics of the sector in the country in question, Bank support to a reform process might be more effective when designed as an investment operation at the beginning of the process to set up the necessary conditions to enable policy implementation, followed by disbursements of resources linked to the achievement of specific goals through policy change. Part Two Separation of functions 8.9 A balanced depth and parallel pace of required legal changes and necessary technical developments should be always maintained in a reform process. When a specific role is to be assigned to an institution in a health sector reform program, it is not enough to secure the legal changes--equal emphasis has to be placed on the technical implementation of the reforms. As an example and in retrospect, the design of organizational changes was initially viewed as a process that was independent of the pace of decentralization to CCSS providers. The original project design did not include the approval of a law on decentralization or any other legal changes. As it became clear that the decentralization process would require a longer implementation period, the expectations regarding the implementation of organizational - 24 - changes should have also been delayed. Another example is the design of the separation of functions between the CCSS and the MOH. There were no effective mechanisms proposed in the project to put the MOH in a material position to be able to fulfill its assigned policy-making role in the health sector. Project design should have taken into account that it is necessary to put in place a transitional period that requires financing during project life and that the transformation process of the MOH's role takes much longer than expected. The project was dependent on reforms in the MOH which were to be supported by the IDB project and the lack of integration produced discordant results in terms of the pace and depth of reform. 8.10 Institutional reform and development should be tackled with a combination of mechanisms to induce change and should not be misunderstood with activities to enhance human resource capacity to perform within the rigidities of the existing institutional structure. Project design and implementation should approach institutional reform and development almost as an underlying base of all reform aspects but not as a separate project component runs the risk of strengthening an existing set of institutional processes and relationships not compatible with the reform strategies and goals. 8.11 The reform process cannot afford to overlook the power exercised by the appropriate incentive mechanisms in the production of institutional change. A lot can be done if the power of resource reallocation is recognized, often greater and faster than what the implementation of the traditional activities such as human resource training and the development of an institutional reform blueprint can accomplish. 8.12 The implementation of activities supporting the separation of financing and delivery of services need to be revised when facing vertical integration as in the case of the CCSS. A "bottom up" approach introducing management contracts and changing the institutional culture is necessary but not sufficient; strong political will and support must be exercised at all times from both within and outside the institution to overcome opposition from interest groups and to manage the power of labor unions. In this sense the predominant approach followed until now in Costa Rica could be accompanied by "top down" interventions where fundamental change is introduced by defining new institutional structures and clear new functions and responsibilities to be exercised accompanied by appropriate reimbursement policies. 8.13 Strong leadership should be exercised when reform momentum slows. The separation of functions and decentralization has made considerable progress, as part of a gradual implementation process, but major political decisions regarding reorganization of the central CCSS structure have been delayed. Lack of political will and leadership to control the well-identified interest groups opposing the reform process from within the CCSS is the critical reason for this situation. This situation must be reversed if current benefits are to be sustainable and pending issues need to be concluded successfully. 8.14 Clearly defined functions and processes are needed for a swift and smooth decentralization of functions. There exists confusion in the health care vice presidency about the role that is to be played at the central and the regional levels. Functions currently being reassigned to the regional level include the evaluation of the service contracts or "compromisos de gestión" as well as the assignment of financial resources to service providers; clearly, having both functions in the same hands reverses the established separation of purchasing and provision of health care service delivery. The lack of clarity has been a major obstacle to further development of a redesigned organizational structure. 8.15 In order for the separation of functions to promote real efficiency gains, change must be made in the payment mechanism to health care providers. While important changes were introduced through the Decentralization Law and the management agreements, the failure to link provider payments with performance has been a limiting factor in the reforms. Part of this limitation is reflected in the project's limited impact on waiting lists and the persistent inefficiency in the hospital network. Because of these shortcomings (a) the separation of the purchasing and the provision function has not transferred sufficient risk to CCSS providers and CCSS hospital managers do not assume risk or reward for the performance of the hospitals, or business units within hospitals; (b) there is no effective competition among health care - 25 - providers, as the insured population is not allowed to change the assigned provider and those providers whose production level is not sufficient to maintain sustainability are bailed out by the CCSS. Moreover, the bulk of service provision is purchased from within the CCSS network only; and (c) if the evaluation of service contracts with providers specify economic incentives according to compliance with indicators based on institutional goals, the appropriate information system to generate such indicators must be in place and fully operational to support the implementation of contracts. 8.16 The implementation of the DRGs is a complex, highly technical issue that must be accompanied by concomitant changes in financing and information systems and cannot overlook the design of a strong and aggressive strategy for its introduction to potential users. While the project successfully implemented DRGs as tool for clinical management, the effectiveness of the instrument was limited by a number of conditions. First, the supporting hospital information system was not adequately maintained throughout the test period. Second, DRGs are regarded as complicated, not understandable, and difficult to use, indicating that training quality was poor or insufficient. Third, the software was imported from the USA and adapted with Spanish modules; insufficient efforts were made to explain to health staff how the system would be adapted to the reality of Costa Rican hospitals and how the DRGs could be interpreted with CCSS data. Fourth, the introduction and pilot-test of DRGs was regarded as low priority by the CCSS according to interviewees. These perceptions have a negative effect on users. Fifth, DRGs were never integrated with the hospital management system and continues to be a challenge. As a consequence, they are not yet perceived as part of an integral hospital management solution. They should be included as an instrument for decision making in all management training programs and as a powerful mechanism to allocate financial resources. In conclusion, as one interviewee put it, "they became one more system we had to fill out, one of the thousand forms we are obliged to fill out". Part Three Redefining the primary health care model 8.17 In the context of project financing, it is beneficial for the implementation of the new PHC-model to comprehensively pilot-test before embarking on full-scale implementation. Fortunately, the CCSS was able to allocate additional resources to overcome the lack of planning reflected by the national implementation without pilot testing. Several difficulties arose during the process of implementation: (1) incremental salary costs were grossly underestimated as the insured population attended by an EBAIS was reduced from 5000 to approximately 4200 persons. The estimated incremental salary costs increased six times (from US$5.8 million to $36.6 million), total project cost doubled from $32 million at appraisal to $65 million of actual expenditure and under less benevolent circumstances, this change alone could have meant the abrupt termination of the project (see also section 8.2); (2) retraining of health personnel of the primary care level turned out to be slower than expected; (3) The redefinition of the primary level has direct consequences on the demand of services of the secondary and tertiary levels, both in terms of a varying service mix and in quantitative terms. These issues have to be carefully addressed to prevent stressing ad-hoc adjustments especially at the secondary level; (4) The mechanism that effectively coordinates all three service levels is the referral/counterreferral system. This system must be tested and operating simultaneously with the introduction of the new PHC-model; (5) The redefined PHC-model includes an Information System to adequately monitor and evaluate health services, the equipment necessary to provide the assigned services, and the complementary administrative capacity at each health area. Once the optimal PHC-package is defined and tested, it should be implemented altogether as only the combination of its elements guarantee the expected results. 8.18 The implementation of the PHC-model does not ensure cost-reduction if certain additional measures are not taken simultaneously. Two additional measures could be identified as being necessary for the PHC-model to produce positive efficiency effects. First, the entry to the health service network must - 26 - be restricted to the primary service level, with the exception of emergency cases. While the increased capacity at the PHC level has reduced unnecessary hospitalization, many patients continue to go directly to hospitals for care and use emergency services as a `back door' into the system. Second, transparent and clear regulations for the participation of private providers are essential to introduce competition and to thereby effectively foster efficiency in the provision of primary health services. This measure, though included in the project design, was not implemented for unknown reasons. 9. Partner Comments (a) Borrower/implementing agency: Al concluir una década de haber iniciado el proceso de transformación del sector y de modernización de la Caja Costarricense de Seguro Social, con el apoyo financiero y técnico del Banco Mundial, me complace saber que hoy mantenemos firmemente los mismos principios de dirección, cambio y de esta manera conservar las conquistas en materia de salud y seguridad social alcanzadas por nuestro país desde finales del siglo XX. En virtud de lo anterior y dada la importancia que representa el Informe de Cierre, hemos considerado pertinente llevar a cabo una revisión profunda del mismo y confiamos que nuestras observaciones contribuyan hacia el enriquecimiento del documento que ha preparado el Banco. Además de las referencias de evaluaciones citadas en el documento es importante destacar que existen dos importantes informes de evaluación sobre el cumplimiento de metas, objetivos, avances y limitaciones del proceso de reforma del sector salud, que representan fuentes adicionales de referencia y podrían contribuir a robustecer esta sección. Estos son: "Panorama de la Reforma del Sector Salud", 1999, elaborado por la Contraloría General de República de Costa Rica y segundo, "VIII Informe del Estado de la Nación", 2002. En relación con el área de fortalecimiento institucional, se debe destacar la implementación de las juntas de salud y el fortalecimiento del proceso de planificación institucional, con el diseño y elaboración del primer Plan de Atención a la Salud de las Personas (PASP), cuyo aporte central recae en la búsqueda de valor en la inversión en salud mediante la priorización de 19 grupos de problemas de salud, 3 problemas de servicios, 29 intervenciones y 138 metas de reducción del riesgo y de mejora de la oferta de servicios, de formación y capacitación del recurso humano y de necesidades de información e investigación. También se desarrolló el diseño e implementación de la plataforma tecnológica por medio del Sistema Básico de Gestión de Suministros de la CCSS (actividad financiada con recursos del Proyecto). El mismo permitirá que los procesos de planificación, compras, licitaciones y almacenes se encuentren interconectados y actúen en forma interdependiente e interrelacionada para contar con información ágil y oportuna en el proceso de abastecimiento de los artículos con economías de escala. En relación con los Grupos Relacionados al Diagnóstico, es importante notar que el Proyecto de Modernización de la CCSS, ha llevado adelante diversos proyectos encaminados a incorporar nuevas formas organizativas y de gestión en los hospitales, siempre orientados a mejorar la calidad y eficiencia de los servicios. Todos estos esfuerzos resultarían incompletos sin instrumentos de mejora del conocimiento que permitan avanzar en el proceso de toma de - 27 - decisiones sobre datos objetivos. Este es el objetivo del Proyecto de Implantación de un Sistema de Información por Producto y basada en los GDR. La Caja dispone desde hace décadas de datos clínicos de cada paciente hospitalizado, lo que nos ha colocado de hecho en una posición de avanzada respecto a muchos países de nuestro entorno e incluso a países más desarrollados. El gran esfuerzo de recolección y procedimiento de datos en la Caja, es el que ha permitido incorporar de forma fácil, instrumentos de medida de la actividad de los hospitales en términos clínicos. La aparición de Sistemas de Clasificación de Pacientes como expresión de la actividad desarrollada por los hospitales, constituye un cambio muy importante para la información, al permitir referir los indicadores a tipologías específicas de las patologías hospitalarias. Los GDRs han adquirido además una amplia difusión internacional, abriendo todo un mundo de posibilidades de contraste de la información hospitalaria. Desde el inicio del Proyecto en 1997, se han obtenido ya logros importantes disponiendo de toda la información de hospitalización por GDR desde la década de los 90. Esta información se produce además puntualmente en la Unidad de Estadísticas Biomédicas desde 1998 y algunos hospitales de la CCSS disponen también, de los medios para analizar su actividad mediante GDR. La principal aportación de los GDR es mejorar el conocimiento, la calidad y la eficiencia hospitalaria basándose, por primera vez, en los pacientes tratados que constituye la finalidad de los hospitales. En relación con la implantación del proyecto, quisiéramos señalar los siguientes puntos: La unificación de esfuerzos entre el Proyecto de Modernización y la Gerencia de Modernización y Desarrollo, responde a la necesidad de garantizar la sostenibilidad de las innovaciones generadas dentro del marco de la reforma, estableciendo una vía efectiva y eficaz de transferencia de conocimientos al ámbito institucional. En ese momento, no se había estudiado ni visualizado la necesidad de un segundo préstamo con el apoyo del Banco Mundial. Sería importante, en evaluaciones futuras, analizar la posibilidad de incorporar las percepciones de los usuarios internos y externos en el análisis de resultado e impacto de los programas y actividades realizadas con los recursos del Proyecto. Asimismo, es importante mencionar un aspecto muy positivo, referente a que durante la ejecución del Proyecto no se evidenciaron huelgas, paros o mayor resistencia laboral a las propuestas y ejecución de las reformas estructurales y modernización institucional. (En presencia de 29 sindicatos a lo interno de la CCSS). Finalmente, consideramos que en el anexo 6, al pie de página, se debe incluir una definición de los criterios de evaluación de desempeño, a saber: altamente satisfactorio, satisfactorio, insatisfactorio y altamente insatisfactorio, por cuanto es importante que todos los lectores nacionales e internacionales del documento de Informe de Cierre, logren efectuar comparaciones - 28 - con otros préstamos, tanto para el mismo país, como con otros préstamos con financiamiento externo. (b) Cofinanciers: N/A (c) Other partners (NGOs/private sector): N/A 10. Additional Information Bank Team María Luisa Escobar (Lead Health Economist, ICR Task Leader) James Cercone (Health Economist and Health Policy Analyst) Vilma Ibarra (Journalist) Stefan Brunner (Economist) Sonia Levere (Language Program Assistant) Counterpart Team Norma Montero Juan Carlos Sánchez Carlos Montoya Claudio Arce Giovanni Márquez Geovanna Rojas Comments provided from: Evangeline Javier Susana Augusto Helena Ribe Helen Saxenian - 29 - Annex 1. Key Performance Indicators/Log Frame Matrix Outcome / Impact Indicators: 1 Indicator/Matrix Projected in last PSR Actual/Latest Estimate 1. EBAIS HQs with proper cold chain 98% 100% maintenance 2. Discharged after delivery, with > 3 penatal 90% 83% visits 3. Women 45-65 yrs. hospitalized who had 80% 25% VC past two years 4. Hypertensive population, under care 75% 75% 5. EBAIS with medical equip. in good 100% 95% conditions 6. Patients > 15 yrs with record of blood 100% 98% pressure measurement in last consultation 7. Patients satisfied with care received for 80% 80% hypertension 8. Diabetic patrients in area, under medical 75% 65% control 9. Patients satisfied with the care received 80% 86% 10. Patients with MI, who sere seen in 10% max N/A outpatient services for symptoms and not diagnosed correctly at first contact 11. EBAIS physicians that received 80% 100% cardio-pulmonary (CPR) resuscitation courses in last 5 years 12. EBAIS with adequate basic equipment 100% N/A for primary care of MI Output Indicators: 1 Indicator/Matrix Projected in last PSR Actual/Latest Estimate STRENGTHENING THE CCS 1.Design structure, functions, norms and 100% 100% procedures for the three Complexes Review action plan and timetable Approval by CCSS Board of Directors Implement new structure and develop operational norms (% of annual targets) 2.Management Information System (MIS) 100% ·Prepare TORs, select consultants for study, 100% prepare procurement documents ·Design the MIS for each module 70% ·Prepare and review action plan 70% ·Implement action plan (% of annual targets) 0% 3. Develop training program for managers, central and regional staff No. of Courses 2 100% No. of Participants 60 100% REDEFINED PRIMARY HEALTH CARE MODEL 1. Design pilot plan in Huetar Norte 100% 100% Review pilot plan Implementation (% of annual targets) Transfer of staff (% of annual targets) 2. Transfer facilities and personal from MOH 12/31/95 end 100% 100% and implement EBAIS in Central Norte, Central Sur, Huetar Atlántica Implementation (% of annual targets) - 30 - 3. Implement Quality Assurance system in 10/30/94 100% Central Norte, Central Sur, Huetar Atlántica 4. Transfer facilities and personal from MOH 12/31-96 end 100% 100% and implement EBAIS in Chorotega, Pacífico Central, Brunca Implementation (% of annual targets) 5. Implement Quality Assurance system in 10/30/95 100% Chorotega, Pacífico Central, Brunca 6. Prepare work annual plan and budgets for 07/31/98 100% regions 7. Carry out training of EBAIS and 100% |00% supervisors No. of health professionals No. of auxiliary staff No. of new auxiliary staff RESOURCE ALLOCATION AND PILOT TESTING OF ALTERNATIVE COMPONENT - Redesign financing model 100% - Normalized collection of CCSS premiums 100% - Improved administration of financial 100% resources - Improved resource allocation 100% - Development of unit cost and economic 60% evaluation of medical procedures - Evasion reductions 50% N/A - Health coverage for salaried workers 95% 88% - Health coverage for independent workers 95% 25% - Recruitment of senior consultant firm for 100% MIS - MIS testing in 100% a. health area b. one region - All areas 50% INCIENSA - Feasibility study 100% - Activities 100% PROJECT ADMINISTRATION 1. Selection of six professionals to staff the 100% 100% PCU 2. Develop Standard Bidding Documents and 100% 100% contracts for civil works, goods, and consultants 3. Progress in project implementation 100% 100% measured accounting to approved annual work plan targets and budgets (% of annual targets) 4. Loan distributed according to approved 100% 100% work plans and budgets (% of annual targets) 5. Consultants contracted according to 100% 100% scheduled studies and technical assistance (% of annual targets) 6. Procure equipment 100% 100% 1End of project - 31 - Annex 1.b Matrix of Objectives and Outcomes by Component Objective - Outcome Matrix by Component Component General No. Specific Objectives General Outcome Objective I. CCSS Institutional 1 Reorganization of CCSS into four units: · All pension management was separated from the health Development Policy Level (Board of Directors and and disability insurance and cross-subsidization from the Executive President), Health Care pensions vice presidency to the health vice presidency was Assisting the CCSS in Complex, Pensions Complex and Financial eliminated. implementing Mgt. Complex. institutional reforms at 2 Decentralization of functions of the Health· The separation of functions between the CCSS (financing the central level and in Care (7 regional offices), Pensions and and provision) and the MoH (stewardship and regulation) strengthening the Financing complexes. was effectively accomplished, thus eliminating the costly support systems and the 3 Introduction of a Management Information duplicity in personnel and infrastructure. . decentralization process System (MIS) 4 Human Resource Development to · The separation of the financing and health delivery implement reform including senior and functions initiated with introduction of more than 130 middle management and service delivery management contracts between CCSS and provider staff (MIS, planning, procurement, network. personnel administration, and others). 5 Manpower Planning Includes personnel · Law on Decentralization approved in 1998. policies, decentralization, norms & procedures, staffing patterns, personnel · By 2002, 34 (of 133 total) providers operating under incentives, civil service. decentralized framework. 6 Pharmaceutical and Medical Supplies System · MIS to manage pharmaceutical procurement and inventory 7 Rational use of medicines: brand selection, developed and implemented. correct dosage and adequate communication to patients for its use · UCR study on contracting human resources. · The project trained 232 administrative staff and 14,888 course and seminar participants. As a percentage of all management staff, it is estimated that roughly 60 percent of all CCSS management received training under the Project · Five master programs implemented in Health Economics, Public Health Management, Epidemiology, Public Health, and Health Service Management. - 32 - Component General No. Specific Objectives General Outcome Objective II. Redefined Primary 8 Introduction of integrated (preventive · Coverage was assured by creating 102 Health Areas in the Health Care Model, and curative) health care packages for country's territory (98 are already installed or 96%), each of Quality Assurance children, adolescents, adult, which comprises around ten to fourteen basic medical units System, & Hospital Study elderly,and women. called EBAIS. 9 Continuity of service providers Supporting the 10 Introduction of epidemiological · In total 818 EBAIS have been established covering 88% of implementation of a surveillance the country's population. Medical equipment was acquired redefined model of primary 11 Enforcement of referrals and new personnel was hired to implement the new health care services which 12 Support from community organizations PHC-model. would be based on national 13 Efficient PHC provision priority health care · Integrated health care packages for children, adolescents, programs and, at the same 14 Establishment of 74 Health Areas and adults, elderly and women were introduced to directly time, be responsive to 7 Regional Offices address national epidemiological priorities. regional and local 15 Establishment of 800 EBAIS epidemiological 16 Development of a Quality Assurance · The process was accompanied by a shift of budgetary conditions. System for PHC. 17 Survey of secondary and tertiary levels allocations toward the primary health care level, increasing defining functions, staffing, its participation in health insurance expenditure from procedures, control mechanisms, around 15 pre-reform to around 23 percent post-reform. financing alternatives and the role of the private sector to achieve · Quality assurance was included as part of the PHC model cost-effectiveness and quality. and introduced through performance indicators in the management contracts that are established with all Health Areas. - 33 - Component General No. Specific General Outcomes Objective Objectives III. Resource Management 18 Strengthening of the incentives for cost Resource allocation: and Pilot Testing of control and efficiency in health service · Gains were made possible by the introduction of Alternative Health Care delivery management agreements, entitled Compromisos de Financing 19 Introduction of a monitoring and evaluation Gestion, between the Health Services Contracting Strengthening the CCSS information system of health services. Department and the hospitals and health area financial complex by (MIS) providers. In total, 133 contracts signed annually. improving the resource 20 Review methods of cost calculation of · Improvements in equity are evident at the primary management, allocation and health services care level as capitated payment mechanisms were evaluation systems, reducing 21 Reduce evasion by improving the process implemented in 1997. evasions increasing of reporting earnings and collecting · In hospitals a performance based payment system contributory control of social contributions, and increase affiliation was introduced, based on the introduction of a unique security premiums, and pilot 22 Analyze the feasibility of alternative unit of hospital production--the UPH (Unidad de P testing, with the expectation of revenue sources for the CCSS health roduccion Hospitalaria). This production unit wider implementation, of insurance fund allowed the CCSS to link production with payment alternative models of health 23 Eliminate the outstanding Government and replace the historical budget for hospitals with a care financing and hospital debt to the CCSS performance-based budget. administration. 24 Evaluate existing alternative delivery and · Improvements in efficiency are notable both in financial systems terms of the allocation between levels (primary level No. Specific Objectives funds represented around 15 pre-reform to around 23 percent post-reform) and in the technical efficiency with which the hospitals deliver services (average 25 Review alternative health service delivery length of stay decreased more than 10 percent and and financial models (fixed capitation production of discharges and consultations increased transfers and DRG-related payments) 32 percent over the 96-02 period). 26 Pilot-test proposed alternative delivery and · Improvements made to the formulas for calculation financing systems of hospital costs for non-insured. Insurance revenue: · Centralization of all health and pension premium collections through the development and deployment of the universal revenue collection mechanism SICERE (Sistema Centralizado de Recaudación) in 2000. · Short-term effect: estimated savings of US$6 million per year due to reduced process times. · Between 1993 and 2001, the average debt of Government (estimated as the difference between what the Government should pay and what is actually sent to the CCSS) was reduced from over 65 percent to just over 14 percent. · The project also managed to promote the interchange of information between the CCSS, the MOF and the National Insurance Institute (INS) which triangulates reported income between the three institutions and thereby allows inspectors to more effectively target evasion. General Outcomes Pilot-testing of alternatives: Pilot testing was substituted for full implementation of the proposed measures. · Hospitals: DRGs implemented in 24 of 29 hospitals (all acute hospitals operating with DRGs). The DRG implementation process was the first country to fully implement DRGs in Latin America and a significant learning experience was developed in Costa Rica. · Produced significant reductions in average length - 34 - of stay, the number of outlier cases and improved bed and patient management. · CCSS hospitals no longer receive budgets based on historical budgeting but rather based on production and performance targets. · Primary level: The provision of services under a captitation payment mechanism has shown excellent results in comparison with the traditional CCSS model, indicating that the private sector can provide the PHC package of service at a considerable cost-saving in comparison to the traditional, public model. · At present, all health areas receive their annual budget based on the total population, without risk adjustment, eliminating the historical budgeting that characterizes the pre-reform phase. IV. National Health 27 Definition of INCIENSA´s functions, · The laboratory facilities for infectious diseases Surveillance and Quality organizational structure, staffing pattern, priority and febrile and hemorrhagic diseases were Control Laboratory programs, and its relationship with MOH rehabilitated Designing and implementing 28 Identification of current research, teaching and · An integrated information system was feasibility studies for the community activities which would be transferred developed and implemented facilitating development of a National to the University of Costa Rica or other INCIENSA's management of critical information Surveillance and Quality institutions. · INCIENSA staff received considerable training Control Laboratory. 29 Improvement of cost-recovery mechanisms that and technical assistance which improved the allow INCIENSA to retain resources obtained technical capacity of the staff. from service fees. 30 Rehabilitation of laboratories - 35 - Annex 2. Project Costs and Financing Project Cost by Component (in US$ million equivalent) Appraisal Actual/Latest Percentage of Estimate Estimate Appraisal Component US$ million US$ million Institutional Strengthening 9.10 3.68 40.43 Redefined Model of Primary Health Care 17.30 49.95 288.71 Resource Management & Pilot Testing of Alternatives 2.30 8.48 385.4 National Health Surveillance and Quality Control 1.00 0.98 89.89 Laboratory Project Administration Unit 2.30 2.37 102.94 Total Baseline Cost 32.00 65.46 Total Project Costs 32.00 65.46 Total Financing Required 32.00 65.46 Annex 2b. Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) 1 Procurement Method Expenditure Category ICB NCB 2 N.B.F. Total Cost Other 1. Works 0.00 1.50 0.00 0.00 1.50 (0.00) (0.80) (0.00) (0.00) (0.80) 2. Goods 5.30 1.40 0.50 0.00 7.20 (3.60) (1.00) (0.40) (0.00) (5.00) 3. Services 0.00 0.00 16.00 0.00 16.00 Studies & Technical (0.00) (0.00) (15.30) (0.00) (15.30) Assistance, PCU, Training Programs, Fellowships 4. Miscellaneous 0.00 0.00 0.50 0.00 0.50 Printing of Training, (0.00) (0.00) (0.50) (0.00) (0.50) Materials and Manuals 5. Miscellaneous 0.00 0.00 1.00 0.00 1.00 Incremental Operating (0.00) (0.00) (0.40) (0.00) (0.40) Costs 6. Miscellaneous 0.00 0.00 0.00 5.80 5.80 Incremental Salaries for (0.00) (0.00) (0.00) (0.00) (0.00) New Staff Total 5.30 2.90 18.00 5.80 32.00 (3.60) (1.80) (16.60) (0.00) (22.00) - 36 - Annex 2c. Project Costs by Procurement Arrangements (Actual/Latest Estimate) (US$ million equivalent) 1 Procurement Method Expenditure Category ICB NCB 2 N.B.F. Total Cost Other 1. Works 0.00 0.40 0.00 0.00 0.40 (0.00) (0.20) (0.00) (0.00) (0.20) 2. Goods 5.70 3.70 0.70 0.00 10.10 (3.90) (1.40) (0.50) (0.00) (5.80) 3. Services 0.00 0.00 17.00 0.00 17.00 Studies & Technical (0.00) (0.00) (15.30) (0.00) (15.30) Assistance, PCU, Training Programs, Fellowships 4. Miscellaneous 0.00 0.00 0.10 0.00 0.10 Printing of Training, (0.00) (0.00) (0.10) (0.00) (0.10) Materials and Manuals 5. Miscellaneous 0.00 0.00 1.20 0.00 1.20 Incremental Operating (0.00) (0.00) (0.20) (0.00) (0.20) Costs 6. Miscellaneous 0.00 0.00 0.00 36.60 36.60 Incremental Salaries for (0.00) (0.00) (0.00) (0.00) (0.00) New Staff Total 5.70 4.10 19.00 36.60 65.40 (3.90) (1.60) (16.10) (0.00) (21.60) 1/Figures in parenthesis are the amounts to be financed by the Bank Loan. All costs include contingencies. 2/Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local government units. Annex 2d. Project Financing by Component (in US$ million equivalent) Percentage of Appraisal Component Appraisal Estimate Actual/Latest Estimate Bank Govt. CoF. Bank Govt. CoF. Bank Govt. CoF. CCS Institutional Reform 4.60 4.50 9.10 3.20 0.50 3.70 69.6 11.1 40.7 & Development Redefined PHC Model, QA 12.10 5.20 17.30 8.70 41.30 49.90 71.9 794.2 288.4 Resource Management & 1.50 0.70 2.20 7.60 0.90 8.50 506.7 128.6 386.4 Pilot Testing Health Surveillance & 0.90 0.20 1.10 0.70 0.30 1.00 77.8 150.0 90.9 National Laboratory Project Administration 2.00 0.30 2.30 1.40 0.90 2.40 70.0 300.0 104.3 - 37 - Annex 2e. Project Cost by Expenditure Category Reallocation (1) Executed Committed Available Total Available INVESTMENT IBRD CCS Total BRD CCS IBR CCS CCS $ % COSTS S S D S BIRF S A. Civil works 0.26 0.27 0.53 0.19 0.24 0.00 0.00 0.07 0.03 0.10 18% B. Equipment 1. Medical 5.82 3.80 9.62 5.79 3.72 0.00 0.00 0.03 0.08 0.11 Equipment 2. Computer 0.64 0.64 0.00 0.64 0.00 0.00 0.00 0.00 - equipment 0.00 Sub-total 5.82 4.44 10.26 5.79 4.36 0.00 0.00 0.03 0.08 0.10 1% C. Technical Assistance 1. Local 4.41 0.45 4.86 4.36 0.46 0.00 0.00 0.05 -0.01 0.04 2. Foreign 5.97 0.88 6.85 5.94 0.88 0.00 0.00 0.03 0.00 0.04 Sub-total 10.38 1.33 11.71 10.30 1.33 0.00 0.00 0.08 0.00 0.07 1% D. Studies 2.00 0.04 2.04 1.99 0.04 0.00 0.00 0.01 0.00 0.01 1% E. Training 1. Courses & 0.85 0.11 0.96 0.71 0.09 0.00 0.00 0.02 0.02 Seminars 2. Fellowships 2.28 0.46 2.74 2.25 0.29 0.00 0.00 0.00 0.17 0.17 Sub-total 3.13 0.57 3.70 2.96 0.383 0.00 0.00 0.00 0.18 0.18 5% F. Printing material 0.11 0.00 0.11 0.09 0.00 0.00 0.00 0.00 0.00 - 0% 0.00 G. Personnel 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Reassignments TOTAL 21.69 6.65 28.34 21.32 6.36 0.00 0.00 0.18 0.29 0.47 2% INVERSION RECURRENT COSTS A. Incremental Staff 0.00 18.37 18.37 0.00 18.37 0.00 0.00 0.00 0.00 0.00 0% B. Salary 0.00 18.22 18.22 0.00 18.22 0.00 0.00 0.00 0.00 - 0% Differentials 0.00 C. Material and fuel 0.08 0.07 0.15 0.06 0.06 0.00 0.00 0.02 0.01 0.03 17% D. O & M costs 1. Infrastructure 0.00 0.08 0.08 0.00 0.07 0.00 0.00 0.00 0.01 0.01 (O&M) 2 Equipment (O&M) 0.01 0.01 0.02 0.01 0.01 0.00 0.00 0.00 0.00 0.00 Sub-total 0.01 0.09 0.10 0.01 0.08 0.00 0.00 0.00 0.01 0.01 E. Subsistence and 0.02 0.06 0.08 0.01 0.06 0.00 0.00 0.00 0.00 0.00 Transportation F. Other operating 0.20 0.73 0.93 0.16 0.73 0.00 0.00 0.04 0.00 0.04 costs RECURRENT 0.31 37.55 37.85 0.25 37.53 0.00 0.00 0.06 0.01 0.07 0% TOTAL TOTAL PROJECT 22.00 44.19 66.19 21.56 43.89 0.00 0.00 0.44 0.30 0.74 1% - 38 - Annex 2f. Project Costs and Financing by Component and Disbursement Category (in US$ million equivalent) By component and disbursement category COMPONENT AND CATEGORY TOTAL TOTAL Local External SAR Investment INSTITUTIONAL STREGTHENING 9,100,000 3,678,830 450,518 3,228,312 Civil works 0 0 0 Equipment 0 0 0 Consulting Services 3,002,788 272,690 2,730,097 Training and salaried employees 493,927 75,159 418,768 Publications and manuals 3,434 64 3,370 Operating costs 178,682 102,606 76,076 REDEFINED MODEL OF PRIMARY HEALTH CARE 17,300,000 49,947,446 41,289,596 8,657,850 Civil works 0 0 0 Equipment 9,909,835 4,271,068 5,638,766 Consulting services 918,247 131,206 787,041 Training and salaried employees 2,395,489 252,663 2,142,826 Publications and Manuals 43,767 854 42,913 Operating costs 36,680,109 36,633,805 46,304 RESOURCE MANAGEMENT & PILOT TESTING OF 2,300,000 8,478,746 918,508 7,560,238 ALTERNATIVES Civil works 0 0 0 Equipment 0 0 0 Consulting Services 8,033,158 822,081 7,211,077 Training and salaried employees 359,700 43,912 315,789 Publications and manuals 8,230 162 8,068 Operating costs 77,657 52,353 25,304 NATIONAL HEALTH SURVEILLANCE AND QUALITY 1,000,000 988,764 308,312 680,452 CONTROL LABORATORY Civil works 433,295 239,692 193,603 Equipment 189,731 56,941 132,790 Consulting Services 324,850 7,748 317,102 Training and salaried employees 40,888 3,932 36,956 Publications and manuals 0 0 0 Operating costs 0 0 0 PROJECT ADMINISTRATIÓN UNIT 2,300,000 2,367,580 927,318 1,440,262 Civil works 0 0 0 Equipment 52,736 30,751 21,985 Consulting Services 1,383,701 141,033 1,242,668 Training and salaried employees 49,037 7,304 41,733 Printing and manuals 32,418 217 32,201 Operating costs 849,688 748,013 101,675 TOTAL 32,000,000 65,461,366 43,894,252 21,567,114 - 39 - Annex 3. Economic Costs and Benefits Análisis Costo-Beneficio del Proceso de Reforma en Salud Resumen El año 1994 marca el inicio de la reforma en el sector salud costarricense, cuyos objetivos primordiales eran mejorar la equidad en el acceso, lograr la cobertura universal y establecer un nuevo modelo de atención integral que vele por la salud del individuo, la familia y la comunidad. Como parte de una réplica generalizada a nivel internacional, los diversos sectores sociales alegaron que esta reforma no produciría resultados tangibles y mucho menos que llegaría a beneficiar a los grupos más vulnerables. Este trabajo constituye un esfuerzo pionero en la dirección contraria. Por medio de la herramienta de análisis del Costo-Beneficio y realizando estimaciones muy conservadoras de los beneficios asociados a la reforma, demostramos que las conclusiones de aquellos sectores han sido erradas. La readecuación del modelo de atención, el fortalecimiento del papel rector del Ministerio de Salud y la reorganización institucional de la Caja Costarricense del Seguro Social han permitido generar beneficios económicos y sociales que superan con creces las inversiones monetarias realizadas. El valor presente de los beneficios del proyecto suman más que US$89 millones, superando la estimación inicial de US$41 millones de beneficios. Las cifras más relevantes indican que el Valor Presente Neto (VPN) de los beneficios equivale a US$ 86 millones con una Tasa Interna de Retorno (TIR) de 69%.. Aún considerando todas las inversiones del proceso de reforma, US$ 217 millones, los beneficios netos suman US$ 36 millones con una TIR de 36%. Los resultados del presente análisis revelan, sin lugar a duda, que el proceso de reforma costarricense ha arrojado beneficios sustanciales para la sociedad y la economía del país. Introducción Prácticamente todos los países de la región latinoamericana han ejecutado procesos de reforma en sus sistemas de salud, desde inicios de la década del noventa. Costa Rica se unió a este conjunto de países, al iniciar un proceso de reforma en salud en el año 1994, el cual, al igual que en otras latitudes, tuvo diversos opositores. Las principales críticas al proceso tenían como origen la asociación que algunos grupos hicieron entre el proceso de reforma y el concepto de privatización de las instituciones de salud. El argumento que esgrimían estos grupos era precisamente que la reforma a ejecutarse no produciría resultados tangibles y que el bienestar de la población costarricense, finalmente, se vería perjudicado por las acciones que dicho proceso planteaba. Aunque todavía hay retos pendientes de resolver dentro del sector, es claro que sí hubo beneficios tangibles y que estos tuvieron como grupo meta a los usuarios de los servicios de salud y la población en general. Indicadores generales de salud de la población costarricense demuestran que Costa Rica se ubica a la vanguardia de la región latinoamericana en aspectos como su esperanza de vida (77 años), sus tasas de mortalidad infantil (10.8 niños por cada por mil) y cobertura de inmunización para las principales enfermedades prevenibles cercanas al 90%, entre otros. - 40 - Se han realizado diversas publicaciones en las que se muestran los beneficios del proceso de reforma, medidas principalmente a través de indicadores de morbilidad, mortalidad, productividad y gestión. Entre las más destacadas se encuentran algunas ponencias presentadas por la Caja Costarricense del Seguro Social (CCSS) y algunos estudios independientes que plantean que los beneficios en salud que goza Costa Rica son, en parte, consecuencia de las reformas implementadas nueve años atrás. , El principal valor agregado de nuestro estudio es, desde esta perspectiva, la cuantificación en términos monetarios de los beneficios asociados a la reforma en salud y el establecimiento de la rentabilidad que este proyecto presentó durante el periodo 1997-2002. La reforma efectuada se ha financiado por medio de préstamos adquiridos en el exterior, provenientes de fuentes como el Banco Mundial (BM), Banco Interamericano de Desarrollo (BID), Banco Centroamericano de Integración Económica (BCIE), el Gobierno de España y un último proyecto que se gestiona con el gobierno de Finlandia. Cerca de un 19% de la inversión total corresponde al aporte local correspondiente a la CCSS. En el análisis se consideran tres posibles escenarios de análisis, según los montos de los préstamos considerados. En las siguientes secciones se exponen los aspectos conceptuales que permiten comprender a cabalidad la utilidad del Análisis Costo - Beneficio, así como los supuestos establecidos para la realización del presente análisis, la metodología de estimación de los beneficios y las principales conclusiones que se desprenden del mismo. Análisis Costo Beneficio El Análisis Costo-Beneficio (ACB) se basa en un principio bastante sencillo: compara los beneficios y los costos de un proyecto particular y, si los primeros exceden a los segundos, permite contar con un criterio inicial para considerar su aceptabilidad. Los costos y beneficios del proyecto se estiman considerando dos situaciones alternativas: sin proyecto versus con proyecto. Para la comparación de dichas alternativas deben actualizarse los flujos de beneficios y costos en un momento determinado en el tiempo, lo que se logra con la estimación de cifras a valor presente o valor actual . Este último concepto hace referencia al hecho que el dinero tiene un valor en el tiempo y que, tal y como la lógica lo indica, un colón recibido el día de hoy vale mucho más que un colón que se obtenga dentro de cinco años. De esta forma al evaluar debe considerarse el valor de los costos y los beneficios en forma atemporal, es decir en términos de valor presente. Cuando se trabaja con proyectos de carácter social el ACB puede realizarse en forma ex - post, es decir durante la implementación del proyecto o una vez que éste ya haya sido realizado. En estos casos, el ACB es útil para decidir si se continúa con el desarrollo del proyecto o si es viable realizar proyectos adicionales de este mismo tipo. Sin embargo, lo más común es utilizar el ACB en forma ex ante, como una herramienta coadyuvante en la toma de decisiones acerca de la realización o no de un proyecto determinado. El establecimiento de los escenarios "con" y "sin" proyecto pretende comparar la proyección de las tendencias presentes (escenario en el que no ocurre intervención alguna) con las tendencias - 41 - que se producirían como resultado del proyecto. En el caso de la evaluación del proceso de reforma en salud la situación "sin" proyecto implica la proyección de las tendencias en los indicadores previstos tal y como hasta la fecha de inicio del proyecto se venía presentando (ceteris paribus). La situación "con" proyecto está representada por la situación real recopilada por las diversas estadísticas de salud. Dos aspectos primordiales deben tenerse en cuenta al realizar un ACB. En primer lugar, un ACB no es una herramienta contable, sino más bien una herramienta indispensable para la toma de decisiones. En segundo lugar, la realización de un proyecto social, necesariamente implica un costo de oportunidad para la sociedad; es decir, al ser los recursos escasos, su inversión en un proyecto determinado implica que la sociedad renuncia a los beneficios que podrían haberse obtenido en inversiones alternativas. En el ACB que se realiza posteriormente no se contabilizan los costos y beneficios secundarios, también llamados efectos indirectos o externalidades, que se producen como consecuencia del proyecto, pero fuera del ámbito en que este se realiza y sin afectar a su población objetivo. Un ejemplo de estos efectos indirectos lo constituye el caso de las vacunas. Cuando una persona se vacuna contra la gripe, el beneficio directo que se contabiliza es el ahorro en medicinas para la gripe, el salario que dejaría de percibir por no asistir al trabajo para recibir atención médica, etc. Sin embargo, al vacunarse este individuo genera un beneficio adicional en el grupo de personas que frecuenta, pues ellas tampoco se enfermarán, ni pagarán medicinas, entre otros beneficios. Existen tres medidas básicas para la interpretación del ACB: Valor Actual Neto (VAN), Tasa Interna de Retorno (TIR) y relación Beneficio Costo (BC). l VAN De acuerdo con el VAN un proyecto es rentable si el valor actual del flujo de beneficios es mayor que el valor actual del flujo de inversión. En términos más formales un proyecto será rentable si su VAN es mayor que cero y no será conveniente realizarlo si el VAN es menor que cero. l TIR La TIR es la tasa de descuento que convierte en cero el VAN del flujo de beneficios. El criterio de decisión para la TIR implica realizar aquellos proyectos cuyas TIR sean mayores que la tasa de interés que el capital pueda obtener en usos alternativos (costo de oportunidad). Por último la relación BC se construye como el cociente entre el valor actual de los beneficios y el valor actual de los costos. l Relación BC Si la relación BC es mayor que la unidad el proyecto es rentable. Una relación BC igual a la unidad implica un VAN de cero y una BC menor de uno significaría que a la tasa de descuento utilizada, el valor actual de los beneficios es menor que el valor actual de los costos, por lo que no se estaría recuperando la inversión. Una vez establecidas las bases teóricas del ACB procedemos a describir los principales supuestos del análisis, de forma tal que los resultados posteriores sean interpretados desde dicha óptica. - 42 - Objetivos y acciones del proceso de reforma en salud Existen dos tipos de factores que dieron origen al proceso de reforma en salud en Costa Rica. Dentro del grupo que llamaremos aquí de factores endógenos, se encuentran los bajos niveles de productividad de los servicios de salud frente a la creciente proporción de gastos en salud realizados. Definitivamente, el tema de la eficiencia de los servicios de salud fue uno de los principales aspectos incluidos en la reforma propuesta. Además, dentro de la composición por niveles del sector salud, existía una alta inequidad, donde el nivel hospitalario consumía casi un 70% de los recursos del seguro de salud, relegando así la posibilidad de fortalecer el primer nivel de atención. En última instancia, el tipo de asignación de los recursos, de forma histórica, impedía considerar las características sociodemográficas y de morbilidad de las regiones. Como parte de los factores exógenos que impulsaron el cambio se encuentran la transición demográfica que Costa Rica enfrenta y que en el mediano plazo ensanchará la base de la pirámide poblacional. Este futuro mayor porcentaje de personas de la tercera edad requerirán mayores y más complejos servicios de salud, gastos que pueden prolongarse un largo periodo conforme se incrementa la esperanza de vida de este grupo poblacional. La transición epidemiológica constituye el otro factor exógeno, que caracterizará al país por enfermedades prevalentes en los países industrializados. Como respuesta a los retos que enfrentaba el sector salud, el Gobierno y la CCSS impulsaron un proceso de reforma con seis objetivos específicos: i) Mejorar la equidad en el acceso a los servicios de salud ii) Ampliar la cobertura iii) Brindar mayor énfasis a la atención ambulatoria de tipo promocional y preventiva iv) Realizar cambios en la organización y gestión del sector salud v) Impulsar la descentralización y la participación social y comunitaria vi) Incrementar las funciones de conducción política y regulación del Ministerio de Salud así como disminuir su responsabilidad directa como prestador de servicios. El proceso de reforma en salud se financió con organismos internacionales (BM, BID, España y Finlandia) así como recursos propios. Las acciones se concentraron en cuatro componentes principales: 1. Fortalecimiento de la función rectora del Ministerio de Salud, que incluye el desarrollo de cuatro funciones estratégicas: (i) dirección, coordinación y conducción; (ii) vigilancia y planificación de la salud; (iii) regulación, control y evaluación y; (iv) investigación y desarrollo tecnológico. 2. Readecuación del modelo de atención con la creación de más de 800 Equipos Básicos de Atención Integral de la Salud (EBAIS) e inversiones importantes para mejorar la capacidad resolutiva del primer nivel mientras que se ampliaba el acceso. 3. Fortalecimiento institucional de la CCSS para acompañar los procesos de cambio en la prestación de los servicios de salud que implica el nuevo modelo de atención integral e impulso a la descentralización y la separación de funciones. 4. Rediseño de los mecanismos de asignación de recursos para promover mayor equidad y eficiencia en el sistema, vinculando recursos con desempeño. - 43 - Establecemos el ACB como el mecanismo para evaluar el impacto de las reformas mediante las comparaciones entre beneficios y costos. En la siguiente sección se establecen los principales supuestos que fueron utilizados en el ACB para comparar los beneficios con la inversión/costo del proceso de reforma. Supuestos del análisis Debido a que el comportamiento de la realidad no es perfectamente replicable por medio de la utilización de proyecciones y estimaciones, se hace necesario establecer claramente los supuestos sobre los que se sustenta el ACB del proyecto de reforma del sector salud costarricense. Los principales supuestos utilizados son conservadores y no incluyen otros beneficios no tangibles que haya provocado la reforma: l Todas las cifras están en unidades monetarias reales (colones o dólares) del año 2002, por lo tanto los resultados analizados están exentos del efecto inflacionario. l Los beneficios directos están constituidos por la reducción en los siguientes indicadores: días paciente por servicio, tasas de hospitalización, tasa de infecciones intrahospitalarias, egresos para las siete principales causas de morbilidad hospitalaria evitables y tiempo de ejecución del proceso de facturación de la CCSS. l Los beneficios indirectos están representados por el número de años de vida potenciales que el proyecto genera al reducirse las tasas de mortalidad infantil. l Para la obtención de los beneficios directos e indirectos se asumen dos escenarios: sin reforma y con reforma. En el escenario sin reforma el comportamiento de las variables de interés se proyecta asumiendo una tasa de crecimiento similar a la que venía presentando desde inicios de la década de los noventa. El escenario con reforma está constituido por los valores realmente obtenidos y reportados en las estadísticas oficiales de la CCSS. l Se utiliza una tasa de descuento del 10%. l Horizonte temporal del análisis: 1995-2002. l Los beneficios que se generan como consecuencia de las inversiones generadas por el proyecto empiezan a presentarse a partir del año 1997. l Se establecen tres escenarios diferentes en función de las inversiones realizadas en cada uno de ellos: (1) Considera únicamente la inversión realizada por el BM y el aporte local; (2) inversión del BM, aporte local y BID, y (3) BM, aporte local, BID y España. l Debido a que no todos los beneficios son fácilmente cuantificables el presente análisis no toma en cuenta todos los beneficios posibles. Los análisis económicos, que resultan en la valoración económica de la factibilidad de un proyecto pueden de esta forma subvaluar los beneficios resultantes. Por la importancia que implica para el análisis, a continuación se detalla la metodología utilizada para la contabilización de los beneficios en el escenario "sin" reforma. Metodología para la estimación de los indicadores en el escenario sin reforma La estimación de los valores correspondientes a los días paciente, hospitalizaciones e infecciones intrahospitalarias, parte del cálculo de las tasas de crecimiento promedio en el periodo 1990-1995. - 44 - En algunos años las tasas de crecimiento estimadas indican reducciones en algunos indicadores, lo que al compararse con las cifras reales producto de la reforma, indica que el comportamiento daba indicios de mejora, con la reforma esto ocurrió a un ritmo mucho más acelerado. Este es un claro indicador de una mejor gestión hospitalaria y de una gran efectividad del nuevo modelo de atención integral. El ACB distribuye los beneficios sobre el horizonte de vida del proyecto en dos grandes grupos: beneficios directos e indirectos. Los beneficios directos son los beneficios que se obtienen por la reducción en los siguientes indicadores: l número de días paciente por servicio; como consecuencia de una mejor gestión en cada uno de los servicios hospitalarios y el consecuente traslado de casos hacia el primer nivel de atención. l tasas de hospitalización; producto de una mejor atención en los primeres niveles de atención que reduce las hospitalizaciones innecesarias con el consecuente ahorro en costos. l tasa de infecciones intrahospitalarias; frecuentemente se asocia a este indicador con la calidad de los servicios provistos por una institución de salud. La reducción en este caso se contabiliza como el número de días de hospitalización ahorrados más el costo de los medicamentos (antibióticos) provistos para aliviar dichas infecciones. l número de egresos para las siete principales causas de morbilidad hospitalaria evitables (infecciones respiratorias agudas, enfermedad diarreica aguda, control del embarazo, hipertensión arterial, parasitosis intestinal, dermatosis y anemia); en un escenario bastante conservador se estima la reducción de egresos en siete de las principales causas de morbilidad hospitalaria, que producen un ahorro en costos derivado del menor número de estancias. l reducción en el tiempo de ejecución del proceso de facturación de la CCSS a partir de la implantación del SICERE en el año 2001. La estimación se deriva de la siguiente metodología: En los años anteriores al año 2001, cuando entra en vigencia la Ley de Protección al Trabajador, la facturación de la planilla de un mes en particular se realizaba aproximadamente 25 días después de su presentación. Este tiempo se redujo a 14 días a partir de la implementación del SICERE. Esta reducción de tiempo puede valorarse en términos financieros, puesto que los recursos obtenidos en forma temprana pueden ser invertidos en el mercado financiero local (en su mayoría en títulos valores del sector público). Para dicha valoración se utiliza la tasa de interés promedio que reporta el Ministerio de Hacienda para los títulos públicos. l Los beneficios indirectos del proyecto están relacionados con los años de vida potenciales que ahorra el proyecto y el valor financiero y económico de la mayor productividad alcanzada. La valoración económica de las vidas adicionales se realiza calculando el número de años de vida potenciales de cada niño ponderadas por el PIB per cápita costarricense y la tasa de participación en la fuerza laboral de acuerdo a cada rango de edad Luego de haber definido los principales aspectos metodológicos presentamos los principales resultados cuantitativos obtenidos en el ACB realizado. - 45 - Principales resultados En esta sección se consideran tres aspectos esenciales. En primer lugar un análisis de los beneficios obtenidos con el proceso de reforma. En segundo lugar se toman en cuenta los costos de la reforma y finalmente, se presentan los resultados del ACB realizado. Beneficios de la reforma La reforma en salud es, en gran parte, responsable por el alto grado de desarrollo del sector salud. Las ineficiencias que el sector presentaba y los problemas a los que se enfrentaba, fueron atendidas en forma oportuna por medio de los cuatro componentes principales del proceso: fortalecimiento de la función rectora del Ministerio de Salud, readecuación del modelo de atención, fortalecimiento institucional de la CCSS y rediseño de los mecanismos de asignación de los recursos en el sector. Los principales resultados que este proceso produjo en los indicadores de salud que están siendo objeto de análisis se presentan a continuación: 1. Como producto del surgimiento de un nuevo modelo de gestión de los hospitales y el desarrollo efectivo de una red de atención en salud de primer nivel, la reforma en salud tuvo un impacto significativo en la reducción del número de días paciente dentro de las instituciones hospitalarias. Dicha reducción alcanzó una cifra cercana a los 130 mil días, con un beneficio económico reportado de aproximadamente 16 millones de dólares. 2. La reducción en el número de hospitalizaciones a nivel nacional (no solo por principales causas de morbilidad evitable) refleja que la reforma fue capaz de producir un resultado favorable en este indicador. Dichas tasas se redujeron de una cifra cercana al 10% en el año 1990 a una cifra de 8.4% en el año 2002. Esta reducción generó un impacto beneficioso para la CCSS al producirse un ahorro de aproximadamente 63 millones de dólares. 3. La reforma en salud generó una reducción importante en las tasas de infección intrahospitalaria. De una tasa inicial de 16% en el año 1993 se logra obtener una tasa que supera escasamente el 5% en el año 2002. Estas reducciones generan un impacto positivo en la reducción del número de días de estancia así como en el costo por medicamentos (antibióticos) para tratar dichas infecciones. De este modo se estima que la reducción en los días de estancia generó un ahorro de más de 38 mil millones de colones (109 millones de dólares) y la reducción por no consumo de medicamentos alcanzó cifras de 3 mil millones de colones (es decir más de 8 millones de dólares) 4. El impacto económico de la transformación del modelo de atención en salud indica que se evitaron un total de 7,800 hospitalizaciones por causas susceptibles de ser tratadas en el primer nivel de atención. Este ahorro en egresos representó más de 3 mil millones de colones en términos reales (valor presente de más de 2 mil millones de colones), es decir una suma de 9 millones de dólares (6 millones en términos de valor presente). En la Tabla 1 se presentan las reducciones en el número de egresos por categoría, así como el ahorro en costos asociado. - 46 - Tabla1: Reducción en el número de egresos según causas principales de morbilidad y costos asociados Ahorro Reducción egresos Cantidad Ahorro colones Ahorro dólares real real unitario colones Infecciones respiratorias agudas 506 97,318,110 275,274 192,362 Enfermedad diarreica aguda -86 20,780,922 58,781 -240,267 Control del Embarazo 1,926 251,512,277 711,429 130,559 Hipertensión arterial 405 207,901,808 588,072 512,986 Parasitosis intestinal 372 187,956,610 531,655 504,888 Dermatosis 2,198 1,196,280,266 3,383,803 544,356 Anemia 2,467 1,252,454,837 3,542,699 507,776 Total 7,788 3,214,204,829 9,091,713 412,736 Nota: Cifras en colones y dólares están en términos reales (2002=100) El siguiente conjunto de gráficos pretende dar una idea más clara del impacto que la reforma en salud ha generado a partir de la segunda mitad de la década de los noventa. Dicho comportamiento se muestra para las siete causas de morbilidad hospitalaria consideradas en el análisis, las cuales reflejan en mayor medida el fortalecimiento del primer nivel de atención mediante el modelo de áreas de salud. En el panel de gráficos 2a-2g el área con rayas representa el beneficio que la reforma ha generado en términos del número de egresos evitados. - 47 - Panel de Gráficos 2a-2g. Egresos según principales causas de morbilidad evitable - 48 - Nota: El escenario "Sin Reforma" constituye una proyección con las tasas de crecimiento promedio registradas durante los primeros años de la década de los noventa. 5. La reducción en las tasas de mortalidad infantil, producto de la reforma, permitió evitar un total de 919 muertes durante el periodo 1995-2002. Estas muertes evitadas representan una ganancia de 27,500 años de vida ajustados por discapacidad (DALYs por sus siglas en inglés), los que valorados en términos económicos arrojan un beneficio cercano a los 40 millones de dólares. El comportamiento de ambas tasas (real y proyectada) se presenta en el Gráfico 1, a continuación. Gráfico 1 Tasas de mortalidad infantil. 1990-2002 16.0 15.0 14.0 13.0 12.0 11.0 10.0 9.0 8.0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Sin Reforma Con Reforma Nota: El escenario "Sin Reforma" constituye una proyección lineal de las tasas de mortalidad infantil de acuerdo con el comportamiento registrado durante los primeros años de la década de los noventa. 6. La modernización de la que fue objeto el sistema de facturación de la CCSS permitió que se redujera el tiempo que tomaba el proceso. Este pasó de una cifra cercana a los 25 días en el periodo pre-reforma a tan solo 14 días en el periodo 2002-2003. Esta reducción en el número de días genera un ahorro de más de 12 millones de dólares a la CCSS. Costos de la reforma El financiamiento del proceso de reforma en salud corrió por cuenta de varios préstamos realizados con organismos internacionales. En la Tabla 2 se presenta el detalle de los préstamos así como sus respectivas contrapartidas locales. - 49 - Tabla 2: Monto de los préstamos considerados en el ACB BM Aporte local BID Aporte local España BCIE Total 1995 1.11 0.50 0.00 0.56 1.61 3.79 1996 1.89 0.86 0.04 0.21 2.46 5.46 1997 3.91 1.78 2.45 0.08 3.92 12.14 1998 5.1 2.32 1.97 0.11 0.00 9.49 1999 2.25 1.02 1.36 0.30 13.71 0.00 18.64 2000 2.74 1.25 3.47 2.90 21.15 3.90 35.41 2001 2.03 0.92 9.00 3.12 2.60 1.71 19.39 2002 2.97 1.35 4.96 6.01 2.53 1.66 19.47 Total 22.0 10.0 23.2 13.3 40.00 15.27 123.79 La mayoría de estos préstamos fueron aprobados por medio de leyes especiales en la Asamblea Legislativa. El préstamo con el Banco Mundial, fue aprobado en 1994 y los principales componentes del proceso de reforma a los que suministraría fondos fueron la readecuación del modelo de atención, el fortalecimiento institucional de la CCSS, el desarrollo de sistemas alternativos de financiamiento y de administración de los hospitales y el fortalecimiento institucional del Instituto Costarricense de Investigación y Enseñanza en Nutrición y Salud (INCIENSA). El préstamo con el BID para el "Programa de Mejoramiento de los Servicios de Salud" (Ley 7374) se destinó al fortalecimiento del Ministerio de Salud y desarrollo de sus funciones rectoras así como el fortalecimiento de la red de atención primaria en salud (construcción, remodelación y ampliación de puestos y centros de salud posteriormente conocidos como EBAIS) y la construcción del hospital de Alajuela. El préstamo con el Gobierno de España se vinculó con el "Programa Nacional de Renovación del Sistema Hospitalario Nacional", Ley 7683 del 20 de agosto de 1997. Mediante este programa se llevaría a cabo la adquisición de equipos (médicos y no médicos) para distribuir entre las diferentes instituciones de salud. De los US$ 40 millones que constituyen el préstamo US$ 20 millones constituyen un aporte del Instituto de Crédito Oficial del Reino de España y los restantes US$ 20 millones, se adquirieron mediante un préstamo comercial con el Banco Bilbao Vizcaya S.A. Finalmente, se han ejecutado dos préstamos con el BCIE que ascienden a una cifra de US$ 15.3 millones (desembolsados hasta la fecha) y cuyos objetivos principales son la ampliación y construcción de hospitales así como un centro de especialidades oftalmológicas. A pesar que la reforma fue financiada en su mayor parte por estos préstamos, esto no quiere decir que la CCSS no haya realizado un esfuerzo significativo en este proceso, lo cual se ve reflejado necesariamente en términos de gastos, principalmente por las mayores actividades desarrolladas en el nivel de atención primaria. Como parte de nuestra metodología consideramos apropiado valorar los gastos realizados por la CCSS en este periodo de tiempo. Para construir esta serie, se asume que el gasto correspondiente es aproximadamente de un 5% del gasto en salud de la CCSS. Estas cifras representaron para el periodo 1997-2002 un total de 221 millones de dólares en términos reales. - 50 - Todas estas inversiones generaron un sinnúmero de beneficios para la población costarricense. En la siguiente sección se realiza una valoración económica de tales beneficios con el fin de estimar la rentabilidad económica que generó el proyecto de reforma en salud costarricense. Resultados del Análisis Costo Beneficio Aspectos generales En la Tabla 3 se muestran los principales resultados obtenidos para cada uno de los tres escenarios planteados. El escenario número 1 es el más optimista de los tres, donde se considera que los beneficios obtenidos se producen como consecuencia únicamente de las inversiones del BM, el aporte local y el costo estimado para la CCSS por la implementación del nuevo modelo de atención. Bajo estas circunstancias el valor presente de los beneficios totales alcanza una cifra superior a los 30 mil millones de colones (86 millones de dólares), con una TIR de 69%. La TIR obtenida supera con creces la tasa de descuento utilizada en el análisis (10%) y cualquier tasa alternativa de rentabilidad del mercado financiero. La razón beneficio - costo que se presenta para este escenario indica que por cada colón invertido el proyecto de reforma en salud generó aproximadamente 1.5 colones de beneficios, una cifra que corrobora las conclusiones obtenidas a partir de la TIR. Tabla 3: Resumen de resultados Categoría Base Escenario 1 Escenario 2 Escenario 3 VP Beneficios Colones 89,656 Dólares 254 VP Costos Colones 59,252 67,337 76,864 Dólares 168 190 217 VP de los beneficios netos 30,403 22,319 12,792 (millones colones) VP de los beneficios netos 86 63 36 (millones dólares) TIR 69% 53% 36% Razón Beneficio Costo 1.5 1.3 1.2 Nota: El Valor Presente Neto (VPN) de los Beneficios es igual a los beneficios directos e indirectos menos los costos totales del proyecto. La razón Beneficio Costo es igual al valor presente de los beneficios totales dividido por el valor presente de los costos totales. En el escenario 2 que incluye adicionalmente la inversión del BID, el valor presente de los beneficios netos alcanza una cifra de 22 mil millones de colones (63 millones de dólares) con una TIR de 53%. Estas cifras conducen a la obtención de una razón beneficio - costo de 1.3:1, tal y como se muestra en la Tabla 3. En este caso por cada colón que se invirtió en el proceso de reforma de salud se generaron 1.3 colones de beneficio. - 51 - Finalmente, en el escenario 3 se considera el mayor grupo de inversiones (BM, aporte local, BID y España) junto con los costos estimados para la CCSS de la implementación del modelo de atención. En este escenario el valor presente de los beneficios netos alcanza una suma de 12 mil millones de colones (36 millones de dólares), con una TIR de 36% y una razón beneficio - costo de 1.2:1. Bajo cualquiera de los escenarios planteados el análisis demuestra que los beneficios obtenidos por medio del proyecto de reforma en salud superan sin lugar a dudas los costos requeridos para su operación. Adicionalmente debe considerarse que los escenarios de obtención de beneficios considerados son en realidad conservadores, por lo que estas cifras de beneficios en realidad pueden ser aún mayores. En la siguiente sección se considera el comportamiento de los indicadores de rentabilidad previamente analizados, pero utilizando diferentes tasas de descuento y asumiendo una reducción de los beneficios del 10%. De esta manera se plantea qué tan sólidas pueden ser las conclusiones previas, respecto a los beneficios económicos generados a partir de la implementación del proyecto de reforma en salud. A modo de resumen, en la Tabla 4 se presenta una descripción del impacto que la reforma en salud generó para cada una de las variables de análisis y se realiza su valoración en términos económicos. Tabla 4: Impacto económico del proceso de reforma en salud Ahorros Valor presente Resultado Cantidad Colones Dólares Colones Dólares Reducción días paciente 129,971 8,334,588,181 23,575,252 5,616,753,870 15,887,574 Reducción hospitalizaciones 100,869 37,595,317,891 106,342,276 22,276,636,220 63,011,788 Reducción infecciones intrahospitalarias 37,034 67,745,457,135 191,625,089 41,694,064,840 117,936,009 Reducción egresos según 7 causas 7,788 3,214,204,829 9,091,713 2,121,365,080 6,000,502 Reducción en días del tiempo para proceso de facturación (por mes) 11 4,703,245,431 13,303,620 4,479,088,319 12,669,568 Reducción muertes infantiles 919 23,096,648,477 65,331,278 14,120,634,599 39,941,687 Sensibilidad del proyecto El análisis de sensibilidad como su nombre lo indica permite ver qué tan sensibles son los indicadores de rentabilidad del proyecto si se consideran diferentes tasas de descuento para los flujos del proyecto así como posibles reducciones en el monto de beneficios económicos percibidos. - 52 - Los supuestos establecidos en esta sección permiten analizar qué tan rentable hubiese sido el proyecto de haberse utilizado tasas de descuento del 5 y 15%. La tasa del 5% es una de las tasas recomendadas por la Organización Mundial de la Salud para la evaluación de proyectos de corte social tal y como es el caso de nuestro proyecto. Los principales resultados para cada uno de estos escenarios se presentan en la Tabla 5. Tabla 5: Análisis de sensibilidad con diferentes tasas de descuento 5% 10% 15% Valor presente neto beneficios (colones) Escenario 1 41,345 30,403 22,513 Escenario 2 31,052 22,319 16,050 Escenario 3 19,150 12,792 8,340 Valor presente neto beneficios (dólares) Escenario 1 117 86 64 Escenario 2 88 63 45 Escenario 3 54 36 24 De acuerdo con la información provista por la Tabla 5 es claro que los beneficios del proyecto son robustos en los escenarios considerados. Con una tasa del 5% los beneficios del proyecto se incrementarían sustancialmente, alcanzando sumas de más de 41 mil millones de colones (US$ 117 millones). En el caso que la tasa de descuento utilizada hubiese sido superior a la realmente considerada en el análisis, los beneficios presencian una reducción que en el escenario con mayores inversiones reporta un beneficio neto del orden de los US$ 24 millones. Tabla 6: Análisis de sensibilidad, reducción de beneficios Reducción 10% Categoría Base beneficios VPN millones colones Escenario 1 30,403 21,438 Escenario 2 22,319 13,353 Escenario 3 12,792 3,826 VPN millones dólares Escenario 1 86 61 Escenario 2 63 38 Escenario 3 36 11 TIR Escenario 1 69% 53% Escenario 2 53% 37% Escenario 3 36% 18% - 53 - Por otra parte, se estiman las consecuencias de una posible reducción del 10% en el monto de los beneficios percibidos por parte del proyecto de reforma. En este caso y tal como se muestra en la Tabla 6, los beneficios sufren una considerable reducción, pero aún así el proyecto no dejaría de ser rentable. La TIR en el escenario que considera las mayores inversiones se reduce a la mitad, pero supera incluso la mayor tasa de descuento utilizada en el análisis de sensibilidad anterior (15%). Todo parece indicar que las mejoras en salud que la población ha recibido y las reducciones en costos innecesarios para la administración de la CCSS, constituyen dos de los más claros ejemplos de lo que implica la ejecución de un buen proceso de reforma del sector salud, cuya meta principal ha sido, y debe seguir siendo, alcanzar el mayor nivel de bienestar de la población costarricense. Conclusiones Los resultados del presente análisis son contundentes con la afirmación de los beneficios del proceso de reforma en Costa Rica. Logramos no solo la documentación de algunos beneficios, en términos económicos, sino también una demostración concluyente que las inversiones vinculadas con el proceso de reforma produjeron beneficios económicos que superan al costo de la reforma en una proporción de 1.5 a1. Las cifras más relevantes indican que el Valor Presente Neto (VPN) de los beneficios equivale a US$ 86 millones con una Tasa Interna de Retorno (TIR) de 69%. Aún en el escenario más conservador y considerando todas las inversiones del proceso de reforma, US$ 217 millones, los beneficios netos suman US$ 36 millones con una TIR de 36%. Los resultados del presente análisis revelan, sin lugar a duda, que el proceso de reforma costarricense ha arrojado beneficios sustanciales para la sociedad y la economía del país. Si bien existe una agenda pendiente importante, las inversiones realizadas hasta la fecha deberían motivar a los tomadores de decisión a continuar con las reformas. - 54 - Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle No. of Persons and Specialty Performance Rating (e.g. 2 Economists, 1 FMS, etc.) Implementation Development Month/Year Count Specialty Progress Objective Identification/Preparation 2/1992 8 TM; ECONOMIST (2); PUBLIC S S HEALTH SPEC.(1); HEALTH ADMIN SPEC. (1); FINANCIAL SPEC. (1); OPERATIONS SPEC. (1); PUBLIC HEALTH SPEC. (1) 5/1992 7 TM; PUBLIC HEALTH (2); S S HEALTH ECONOMIST (2); HEALTH ADMIN. (1); OPERATIONS ANALYST (1) 7/1992 2 HEALTH SPEC. (1); S S FINANCIAL SPEC.(1) 9/1992 2 TM; HEALTH ADMIN SPEC. S S (1); OPERATIONS ANALYST (1) 11/1992 5 TM; HEALTH ADMIN (1); S S ECONOMIST (1); PROCUREMENT (1); MIS SPECIALIST (1) 3/1993 7 TM; ECONOMIST (1); PUBLIC S S HEALTH SPEC. (1); HEALTH ADMIN. (1); OPERATIONS SPEC. (2); IMPLEMENTATION SPEC. (1); Supervision 12/1993 1 PROCUREMENT ASSISTANT HS HS (1) 05/1995 5 HEALTH ADMINISTRATION HS S (1); SR. ECONOMIST (1); PUBLIC HEALTH (2); ECONOMIST (1) 11/1995 6 TASK MANAGER (1); HS HS HEALTH ADMINISTRATION (1); SR. ECONOMIST (1); PUBLIC HEALTH (2); PROJECT FINANCING (1) 05/1996 3 HEALTH SPECIALIST (1); SR. HS HS ECONOMIST (1); PROCUREMENT SPECIALIST (1) 07/1996 4 HEALTH SPECIALIST (1); SR. S S - 55 - ECONOMIST (1); HEALTH FINANCING SPEC. (1); PROCUREMENT SPECIALIST (1) 12/11/1996 3 HEALTH SPECIALIST (1); S S CONSULTANT (1); PROCUREMENT SPECIALIST (1) 02/1998 4 HEALTH SPECIALIST (2); S S OPERATIONS SPECIALIST (1); HEALTH ECONOMIST (1) 11/1998 1 PROCUREMENT ANALYST S S (1) 01/1999 3 TASK MANAGER (1); U U PROCUREMENT SPECIALIST (1); CONSULTANT (1) 04/1999 1 PROCUREMENT OFFICER (1) S S 07/1999 1 PROCUREMENT ANALYST S S (1) 07/1999 7 TASK MANAGER (1); S S PROC/IMPLEM SPECIALIST (1); CONSULTANT (5) 12/2000 2 TASK MANAGER (1); S S PROCUREMENT SPECIALIST (1) 06/2002 2 TASK MANAGER (1); S S PROCUREMENT SPECIALIST (1) ICR 9/2002 3 TM (1); PROCUR SPEC. S S (1); TEAM ASSISTANT (1) (b) Staff: Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ ('000) Identification/Preparation 34.0 355,149 Supervision 115.45 279,059 ICR 8 32,000 Total 299.26 690,208 - 56 - Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating Macro policies H SU M N NA Sector Policies H SU M N NA Physical H SU M N NA Financial H SU M N NA Institutional Development H SU M N NA Environmental H SU M N NA Social Poverty Reduction H SU M N NA Gender H SU M N NA Other (Please specify) H SU M N NA Private sector development H SU M N NA Public sector management H SU M N NA Other (Please specify) H SU M N NA - 57 - Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating Lending HS S U HU Supervision HS S U HU Overall HS S U HU 6.2 Borrower performance Rating Preparation HS S U HU Government implementation performance HS S U HU Implementation agency performance HS S U HU Overall HS S U HU - 58 - Annex 7. List of Supporting Documents A. INSTITUTIONAL REFORM COMPONENT 1. CCSS. Perfil Funcional y Estructural de la CCSS, April 1993. Perfil Funcional y Estructural de la CCSS, April 1993. Componente Fortalecimiento Institucional de la CCSS, January 1993. Componente Sistema de Suministros, May 1993. Sistema de Suministros Para la Atención Integral de las Personas. February 1993. 2. CCSS. Proyecto Rectoría y Fortalecimiento del Ministerio de Salud (Programa de Mejoramiento de Servicios de Salud CR-0120/BID), August 1992. 3. CCSS. Perfil Funcional y Programa de Sensibilización (Formación), September 1992. 4: CCSS. Fondo Nacional de Formación y Capacitación en Salud, May 1993. 5. CCSS. Estructura Organizativa del Centro de Desarrollo Estratégico e Información en Salud y Seguridad Social, April 1993. 6. Carvajal Torres, Luis M. Estudio de Factibilidad Para Privatizar los Servicios de Distribución de Mercaderías en el Departamento Almacenamiento y Distribución CCSS, May 1993. 7. Government of Costa Rica, Ministry of Planning and Economic Policy. Programa de Reforma del Estado, Programa Nacional de Reforma del Sector Salud, March 1992. 8. InterAmerican Development Bank. Programa de Mejoramiento de Servicios de Salud CR-0120/BID, project and loan documents, 1992. 9. Management Sciences for Health. Estudio de Financiamiento, Adquisición y Distribución de Farmacéuticos y Materiales Médicos, October 1992. 10. Muller, Catherine and L. B. Villalobos. Consultoría en el Área de Recursos Humanos, December 1992. B. REDEFINED PRIMARY HEALTH CARE MODEL 11. CCSS. Propuesta de Readecuación del Modelo de Atención, plus annexes, February 1993. Situación Actual del Modelo de Atención, October 1992. Infraestructura de Servicios de Salud por Cantón y Región Programática CCSS-MS, June 1992. 12. CCSS. Costa Rica: Perfil de Morbi-Mortalidad, 1992. 13. García, Ronny. Informe de Consultoría Sobre Gastos por Programa del Ministerio de Salud, October 1992. 14. Grunberg, Marcelo. Informe de Consultoría Sobre Recursos Humanos y Físicos de la CCSS y el Ministerio de Salud, January 1993. 15. Marín, Fernando. Informe de Consultoría Sobre Readecuación del Modelo de Atención, November 1992. C. FINANCIAL MANAGEMENT AND PILOT TESTING OF ALTERNATIVE FINANCING 16. CCSS. Proyecto de Reforma al Sector Salud: Componente Sobre el Financiamiento, April 1993. - 59 - Análisis del Modelo de Financiamiento Actual y Estrategia Para su Rediseño, April 1993. Rediseño del Modelo de Financiamiento del Sector Salud, April 1993. Propuesta Para un Modelo de Financiamiento, September 1992. 17. CCSS. Modelo General de Compra y Evaluación de Alternativas de Gestión y Atención a la Salud: Término de Referencia, December 1992. 18. Fiedler, John L. and F. Rigoli. The Costa Rican Social Security Fund's Alternative Models: A Case Study of the Cooperative-Based, COOPESALUD Pavas Clinic, September 1991. 19. The Wyatt Company. Costa Rica Health Care Reform: Financial Component Report, February 1993. D. STUDY ON DEVELOPMENT OF NATIONAL QUALITY CONTROL LABORATORY 20. CCSS. Fortalecimiento Institucional del Instituto Costarricense de Investigación y Enseñanza en Nutrición y Salud "Dr. Uriel Badilla Fernández", April 1993. E. PROJECT COORDINATION UNIT 21. CCSS. Conceptualización y Estructuración de la Unidad Coordinadora del Proyecto, April 1993. Perfil y Términos de Referencia de los Puestos de la UCP, April 1993. Esquema Básico Para la Administración Financiera, April 1993. Fortalecimiento Institucional de la Fundación FUCODOCSA, April 1993. F. BACKGROUND REPORTS 22. CCSS Unidad Preparatoria. Proyecto de Reforma del Sector Salud, May 1993. 23. CCSS. Informe Evaluativo de la Gestión Administrativa y Programática de la Caja Costarricense de Seguro Social, plus annexes, April 1991. 24. CCSS Presidencia Ejecutiva. La Caja Costarricense de Seguro Social en la Coyuntura de Crisis, March 1991. 25. Presidencia de la República de Costa Rica, Programa Reforma del Estado, Plan Nacional de Reforma del Sector Salud, March 1993. 26. Homedes, Nuria. La Prevención y Promoción de la Salud en Costa Rica, June 1988. 27. Homedes, Nuria et al. Los Recursos Humanos del Sector Salud en Costa Rica, April 1988. 28. Saenz, Luis B. And M. Leon. Gastos de los Hogares en Servicios de Salud Privados en Costa Rica Durante 1987-1988, 1992. 29. Sanguinetty, Jorge A. Informe de Consultoría Sobre La Salud y el Seguro Social en Costa Rica, August 1988. 30. Vallejo, Cesar and R. Iunes. The Health Sector in Costa Rica: Financing and Efficiency, November 1991. 31. Quirós, Roberto C. La Atención Médica en la Seguridad Social: Métodos Para Mejorar la Relación Costo/Eficiencia de los Programas, October 1991. 32. Vargas, Telmo. Eficiencia del Sector Público y Privatización en Costa Rica, Marcha 1990. - 60 - Additional Annex 8. Supporting Data and Tables Table 8.1 Trends in Government Debt towards the CCSS Cifras reales en US$ Concepto 1997 1998 1999 2000 2001 2002 Deuda consolidada seguro salud 31,579,536 33,545,145 22,946,774 25,062,900 33,978,001 19,949,171 Deuda consolidada seguro pensiones 220,325 -529,475 7,044,127 3,753,986 8,643,923 5,290,837 Deuda consolidada total 31,799,862 33,015,669 29,990,900 28,816,886 42,621,925 25,240,003 Table 8.2. Overall Debt Ratios - 61 - - 62 - - 63 -